added ML training projects in pytorch

$tutor$

@@ -0,0 +1,970 @@
+===============================================================================
+=    W e l c o m e   t o   t h e   V I M   T u t o r    -    Version 1.7      =
+===============================================================================
+
+     Vim is a very powerful editor that has many commands, too many to
+     explain in a tutor such as this.  This tutor is designed to describe
+     enough of the commands that you will be able to easily use Vim as
+     an all-purpose editor.
+
+     The approximate time required to complete the tutor is 25-30 minutes,
+     depending upon how much time is spent with experimentation.
+
+     ATTENTION:
+     The commands in the lessons will modify the text.  Make a copy of this
+     file to practice on (if you started "vimtutor" this is already a copy).
+
+     It is important to remember that this tutor is set up to teach by
+     use.  That means that you need to execute the commands to learn them
+     properly.  If you only read the text, you will forget the commands!
+
+     Now, make sure that your Caps-Lock key is NOT depressed and press
+     the   j   key enough times to move the cursor so that lesson 1.1
+     completely fills the screen.
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 1.1:  MOVING THE CURSOR
+
+
+   ** To move the cursor, press the h,j,k,l keys as indicated. **
+	     ^
+	     k		    Hint:  The h key is at the left and moves left.
+       < h	 l >		   The l key is at the right and moves right.
+	     j			   The j key looks like a down arrow.
+	     v
+  1. Move the cursor around the screen until you are comfortable.
+
+  2. Hold down the down key (j) until it repeats.
+     Now you know how to move to the next lesson.
+
+  3. Using the down key, move to lesson 1.2.
+
+NOTE: If you are ever unsure about something you typed, press <ESC> to place
+      you in Normal mode.  Then retype the command you wanted.
+
+NOTE: The cursor keys should also work.  But using hjkl you will be able to
+      move around much faster, once you get used to it.  Really!
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			    Lesson 1.2: EXITING VIM
+
+
+  !! NOTE: Before executing any of the steps below, read this entire lesson!!
+
+  1. Press the <ESC> key (to make sure you are in Normal mode).
+
+  2. Type:	:q! <ENTER>.
+     This exits the editor, DISCARDING any changes you have made.
+
+  3. Get back here by executing the command that got you into this tutor. That
+     might be:  vimtutor <ENTER>
+
+  4. If you have these steps memorized and are confident, execute steps
+     1 through 3 to exit and re-enter the editor.
+
+NOTE:  :q! <ENTER>  discards any changes you made.  In a few lessons you
+       will learn how to save the changes to a file.
+
+  5. Move the cursor down to lesson 1.3.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 1.3: TEXT EDITING - DELETION
+
+
+	   ** Press  x  to delete the character under the cursor. **
+
+  1. Move the cursor to the line below marked --->.
+
+  2. To fix the errors, move the cursor until it is on top of the
+     character to be deleted.
+
+  3. Press the	x  key to delete the unwanted character.
+
+  4. Repeat steps 2 through 4 until the sentence is correct.
+
+---> The ccow jumpedd ovverr thhe mooon.
+
+  5. Now that the line is correct, go on to lesson 1.4.
+
+NOTE: As you go through this tutor, do not try to memorize, learn by usage.
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		      Lesson 1.4: TEXT EDITING - INSERTION
+
+
+			** Press  i  to insert text. **
+
+  1. Move the cursor to the first line below marked --->.
+
+  2. To make the first line the same as the second, move the cursor on top
+     of the character BEFORE which the text is to be inserted.
+
+  3. Press  i  and type in the necessary additions.
+
+  4. As each error is fixed press <ESC> to return to Normal mode.
+     Repeat steps 2 through 4 to correct the sentence.
+
+---> There is text misng this .
+---> There is some text missing from this line.
+
+  5. When you are comfortable inserting text move to lesson 1.5.
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 1.5: TEXT EDITING - APPENDING
+
+
+			** Press  A  to append text. **
+
+  1. Move the cursor to the first line below marked --->.
+     It does not matter on what character the cursor is in that line.
+
+  2. Press  A  and type in the necessary additions.
+
+  3. As the text has been appended press <ESC> to return to Normal mode.
+
+  4. Move the cursor to the second line marked ---> and repeat
+     steps 2 and 3 to correct this sentence.
+
+---> There is some text missing from th
+     There is some text missing from this line.
+---> There is also some text miss
+     There is also some text missing here.
+
+  5. When you are comfortable appending text move to lesson 1.6.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 1.6: EDITING A FILE
+
+		    ** Use  :wq  to save a file and exit. **
+
+  !! NOTE: Before executing any of the steps below, read this entire lesson!!
+
+  1. Exit this tutor as you did in lesson 1.2:  :q!
+     Or, if you have access to another terminal, do the following there.
+
+  2. At the shell prompt type this command:  vim tutor <ENTER>
+     'vim' is the command to start the Vim editor, 'tutor' is the name of the
+     file you wish to edit.  Use a file that may be changed.
+
+  3. Insert and delete text as you learned in the previous lessons.
+
+  4. Save the file with changes and exit Vim with:  :wq <ENTER>
+
+  5. If you have quit vimtutor in step 1 restart the vimtutor and move down to
+     the following summary.
+
+  6. After reading the above steps and understanding them: do it.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 1 SUMMARY
+
+
+  1. The cursor is moved using either the arrow keys or the hjkl keys.
+	 h (left)	j (down)       k (up)	    l (right)
+
+  2. To start Vim from the shell prompt type:  vim FILENAME <ENTER>
+
+  3. To exit Vim type:	   <ESC>   :q!	 <ENTER>  to trash all changes.
+	     OR type:	   <ESC>   :wq	 <ENTER>  to save the changes.
+
+  4. To delete the character at the cursor type:  x
+
+  5. To insert or append text type:
+	 i   type inserted text   <ESC>		insert before the cursor
+	 A   type appended text   <ESC>         append after the line
+
+NOTE: Pressing <ESC> will place you in Normal mode or will cancel
+      an unwanted and partially completed command.
+
+Now continue with lesson 2.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 2.1: DELETION COMMANDS
+
+
+		       ** Type  dw  to delete a word. **
+
+  1. Press  <ESC>  to make sure you are in Normal mode.
+
+  2. Move the cursor to the line below marked --->.
+
+  3. Move the cursor to the beginning of a word that needs to be deleted.
+
+  4. Type   dw	 to make the word disappear.
+
+  NOTE: The letter  d  will appear on the last line of the screen as you type
+	it.  Vim is waiting for you to type  w .  If you see another character
+	than  d  you typed something wrong; press  <ESC>  and start over.
+
+---> There are a some words fun that don't belong paper in this sentence.
+
+  5. Repeat steps 3 and 4 until the sentence is correct and go to lesson 2.2.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		      Lesson 2.2: MORE DELETION COMMANDS
+
+
+	   ** Type  d$	to delete to the end of the line. **
+
+  1. Press  <ESC>  to make sure you are in Normal mode.
+
+  2. Move the cursor to the line below marked --->.
+
+  3. Move the cursor to the end of the correct line (AFTER the first . ).
+
+  4. Type    d$    to delete to the end of the line.
+
+---> Somebody typed the end of this line twice. end of this line twice.
+
+
+  5. Move on to lesson 2.3 to understand what is happening.
+
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 2.3: ON OPERATORS AND MOTIONS
+
+
+  Many commands that change text are made from an operator and a motion.
+  The format for a delete command with the  d  delete operator is as follows:
+
+  	d   motion
+
+  Where:
+    d      - is the delete operator.
+    motion - is what the operator will operate on (listed below).
+
+  A short list of motions:
+    w - until the start of the next word, EXCLUDING its first character.
+    e - to the end of the current word, INCLUDING the last character.
+    $ - to the end of the line, INCLUDING the last character.
+
+  Thus typing  de  will delete from the cursor to the end of the word.
+
+NOTE:  Pressing just the motion while in Normal mode without an operator will
+       move the cursor as specified.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 2.4: USING A COUNT FOR A MOTION
+
+
+   ** Typing a number before a motion repeats it that many times. **
+
+  1. Move the cursor to the start of the line below marked --->.
+
+  2. Type  2w  to move the cursor two words forward.
+
+  3. Type  3e  to move the cursor to the end of the third word forward.
+
+  4. Type  0  (zero) to move to the start of the line.
+
+  5. Repeat steps 2 and 3 with different numbers.
+
+---> This is just a line with words you can move around in.
+
+  6. Move on to lesson 2.5.
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		     Lesson 2.5: USING A COUNT TO DELETE MORE
+
+
+   ** Typing a number with an operator repeats it that many times. **
+
+  In the combination of the delete operator and a motion mentioned above you
+  insert a count before the motion to delete more:
+	 d   number   motion
+
+  1. Move the cursor to the first UPPER CASE word in the line marked --->.
+
+  2. Type  d2w  to delete the two UPPER CASE words.
+
+  3. Repeat steps 1 and 2 with a different count to delete the consecutive
+     UPPER CASE words with one command.
+
+--->  this ABC DE line FGHI JK LMN OP of words is Q RS TUV cleaned up.
+
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			 Lesson 2.6: OPERATING ON LINES
+
+
+		   ** Type  dd   to delete a whole line. **
+
+  Due to the frequency of whole line deletion, the designers of Vi decided
+  it would be easier to simply type two d's to delete a line.
+
+  1. Move the cursor to the second line in the phrase below.
+  2. Type  dd  to delete the line.
+  3. Now move to the fourth line.
+  4. Type   2dd   to delete two lines.
+
+--->  1)  Roses are red,
+--->  2)  Mud is fun,
+--->  3)  Violets are blue,
+--->  4)  I have a car,
+--->  5)  Clocks tell time,
+--->  6)  Sugar is sweet
+--->  7)  And so are you.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			 Lesson 2.7: THE UNDO COMMAND
+
+
+   ** Press  u	to undo the last commands,   U  to fix a whole line. **
+
+  1. Move the cursor to the line below marked ---> and place it on the
+     first error.
+  2. Type  x  to delete the first unwanted character.
+  3. Now type  u  to undo the last command executed.
+  4. This time fix all the errors on the line using the  x  command.
+  5. Now type a capital  U  to return the line to its original state.
+  6. Now type  u  a few times to undo the  U  and preceding commands.
+  7. Now type CTRL-R (keeping CTRL key pressed while hitting R) a few times
+     to redo the commands (undo the undo's).
+
+---> Fiix the errors oon thhis line and reeplace them witth undo.
+
+  8. These are very useful commands.  Now move on to the lesson 2 Summary.
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 2 SUMMARY
+
+
+  1. To delete from the cursor up to the next word type:    dw
+  2. To delete from the cursor to the end of a line type:    d$
+  3. To delete a whole line type:    dd
+
+  4. To repeat a motion prepend it with a number:   2w
+  5. The format for a change command is:
+               operator   [number]   motion
+     where:
+       operator - is what to do, such as  d  for delete
+       [number] - is an optional count to repeat the motion
+       motion   - moves over the text to operate on, such as  w (word),
+		  $ (to the end of line), etc.
+
+  6. To move to the start of the line use a zero:  0
+
+  7. To undo previous actions, type: 	       u  (lowercase u)
+     To undo all the changes on a line, type:  U  (capital U)
+     To undo the undo's, type:		       CTRL-R
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			 Lesson 3.1: THE PUT COMMAND
+
+
+       ** Type	p  to put previously deleted text after the cursor. **
+
+  1. Move the cursor to the first line below marked --->.
+
+  2. Type  dd  to delete the line and store it in a Vim register.
+
+  3. Move the cursor to the c) line, ABOVE where the deleted line should go.
+
+  4. Type   p   to put the line below the cursor.
+
+  5. Repeat steps 2 through 4 to put all the lines in correct order.
+
+---> d) Can you learn too?
+---> b) Violets are blue,
+---> c) Intelligence is learned,
+---> a) Roses are red,
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		       Lesson 3.2: THE REPLACE COMMAND
+
+
+       ** Type  rx  to replace the character at the cursor with  x . **
+
+  1. Move the cursor to the first line below marked --->.
+
+  2. Move the cursor so that it is on top of the first error.
+
+  3. Type   r	and then the character which should be there.
+
+  4. Repeat steps 2 and 3 until the first line is equal to the second one.
+
+--->  Whan this lime was tuoed in, someone presswd some wrojg keys!
+--->  When this line was typed in, someone pressed some wrong keys!
+
+  5. Now move on to lesson 3.3.
+
+NOTE: Remember that you should be learning by doing, not memorization.
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 3.3: THE CHANGE OPERATOR
+
+
+	   ** To change until the end of a word, type  ce . **
+
+  1. Move the cursor to the first line below marked --->.
+
+  2. Place the cursor on the  u  in  lubw.
+
+  3. Type  ce  and the correct word (in this case, type  ine ).
+
+  4. Press <ESC> and move to the next character that needs to be changed.
+
+  5. Repeat steps 3 and 4 until the first sentence is the same as the second.
+
+---> This lubw has a few wptfd that mrrf changing usf the change operator.
+---> This line has a few words that need changing using the change operator.
+
+Notice that  ce  deletes the word and places you in Insert mode.
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		       Lesson 3.4: MORE CHANGES USING c
+
+
+     ** The change operator is used with the same motions as delete. **
+
+  1. The change operator works in the same way as delete.  The format is:
+
+         c    [number]   motion
+
+  2. The motions are the same, such as   w (word) and  $ (end of line).
+
+  3. Move the cursor to the first line below marked --->.
+
+  4. Move the cursor to the first error.
+
+  5. Type  c$  and type the rest of the line like the second and press <ESC>.
+
+---> The end of this line needs some help to make it like the second.
+---> The end of this line needs to be corrected using the  c$  command.
+
+NOTE:  You can use the Backspace key to correct mistakes while typing.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 3 SUMMARY
+
+
+  1. To put back text that has just been deleted, type   p .  This puts the
+     deleted text AFTER the cursor (if a line was deleted it will go on the
+     line below the cursor).
+
+  2. To replace the character under the cursor, type   r   and then the
+     character you want to have there.
+
+  3. The change operator allows you to change from the cursor to where the
+     motion takes you.  eg. Type  ce  to change from the cursor to the end of
+     the word,  c$  to change to the end of a line.
+
+  4. The format for change is:
+
+	 c   [number]   motion
+
+Now go on to the next lesson.
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		  Lesson 4.1: CURSOR LOCATION AND FILE STATUS
+
+  ** Type CTRL-G to show your location in the file and the file status.
+     Type  G  to move to a line in the file. **
+
+  NOTE: Read this entire lesson before executing any of the steps!!
+
+  1. Hold down the Ctrl key and press  g .  We call this CTRL-G.
+     A message will appear at the bottom of the page with the filename and the
+     position in the file.  Remember the line number for Step 3.
+
+NOTE:  You may see the cursor position in the lower right corner of the screen
+       This happens when the 'ruler' option is set (see  :help 'ruler'  )
+
+  2. Press  G  to move you to the bottom of the file.
+     Type  gg  to move you to the start of the file.
+
+  3. Type the number of the line you were on and then  G .  This will
+     return you to the line you were on when you first pressed CTRL-G.
+
+  4. If you feel confident to do this, execute steps 1 through 3.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 4.2: THE SEARCH COMMAND
+
+
+     ** Type  /  followed by a phrase to search for the phrase. **
+
+  1. In Normal mode type the  /  character.  Notice that it and the cursor
+     appear at the bottom of the screen as with the  :	command.
+
+  2. Now type 'errroor' <ENTER>.  This is the word you want to search for.
+
+  3. To search for the same phrase again, simply type  n .
+     To search for the same phrase in the opposite direction, type  N .
+
+  4. To search for a phrase in the backward direction, use  ?  instead of  / .
+
+  5. To go back to where you came from press  CTRL-O  (Keep Ctrl down while
+     pressing the letter o).  Repeat to go back further.  CTRL-I goes forward.
+
+--->  "errroor" is not the way to spell error;  errroor is an error.
+NOTE: When the search reaches the end of the file it will continue at the
+      start, unless the 'wrapscan' option has been reset.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		   Lesson 4.3: MATCHING PARENTHESES SEARCH
+
+
+	      ** Type  %  to find a matching ),], or } . **
+
+  1. Place the cursor on any (, [, or { in the line below marked --->.
+
+  2. Now type the  %  character.
+
+  3. The cursor will move to the matching parenthesis or bracket.
+
+  4. Type  %  to move the cursor to the other matching bracket.
+
+  5. Move the cursor to another (,),[,],{ or } and see what  %  does.
+
+---> This ( is a test line with ('s, ['s ] and {'s } in it. ))
+
+
+NOTE: This is very useful in debugging a program with unmatched parentheses!
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		      Lesson 4.4: THE SUBSTITUTE COMMAND
+
+
+	** Type  :s/old/new/g  to substitute 'new' for 'old'. **
+
+  1. Move the cursor to the line below marked --->.
+
+  2. Type  :s/thee/the <ENTER>  .  Note that this command only changes the
+     first occurrence of "thee" in the line.
+
+  3. Now type  :s/thee/the/g .  Adding the  g  flag means to substitute
+     globally in the line, change all occurrences of "thee" in the line.
+
+---> thee best time to see thee flowers is in thee spring.
+
+  4. To change every occurrence of a character string between two lines,
+     type   :#,#s/old/new/g    where #,# are the line numbers of the range
+                               of lines where the substitution is to be done.
+     Type   :%s/old/new/g      to change every occurrence in the whole file.
+     Type   :%s/old/new/gc     to find every occurrence in the whole file,
+     			       with a prompt whether to substitute or not.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 4 SUMMARY
+
+
+  1. CTRL-G  displays your location in the file and the file status.
+             G  moves to the end of the file.
+     number  G  moves to that line number.
+            gg  moves to the first line.
+
+  2. Typing  /	followed by a phrase searches FORWARD for the phrase.
+     Typing  ?	followed by a phrase searches BACKWARD for the phrase.
+     After a search type  n  to find the next occurrence in the same direction
+     or  N  to search in the opposite direction.
+     CTRL-O takes you back to older positions, CTRL-I to newer positions.
+
+  3. Typing  %	while the cursor is on a (,),[,],{, or } goes to its match.
+
+  4. To substitute new for the first old in a line type    :s/old/new
+     To substitute new for all 'old's on a line type	   :s/old/new/g
+     To substitute phrases between two line #'s type	   :#,#s/old/new/g
+     To substitute all occurrences in the file type	   :%s/old/new/g
+     To ask for confirmation each time add 'c'		   :%s/old/new/gc
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		Lesson 5.1: HOW TO EXECUTE AN EXTERNAL COMMAND
+
+
+   ** Type  :!	followed by an external command to execute that command. **
+
+  1. Type the familiar command	:  to set the cursor at the bottom of the
+     screen.  This allows you to enter a command-line command.
+
+  2. Now type the  !  (exclamation point) character.  This allows you to
+     execute any external shell command.
+
+  3. As an example type   ls   following the ! and then hit <ENTER>.  This
+     will show you a listing of your directory, just as if you were at the
+     shell prompt.  Or use  :!dir  if ls doesn't work.
+
+NOTE:  It is possible to execute any external command this way, also with
+       arguments.
+
+NOTE:  All  :  commands must be finished by hitting <ENTER>
+       From here on we will not always mention it.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		      Lesson 5.2: MORE ON WRITING FILES
+
+
+     ** To save the changes made to the text, type  :w FILENAME  **
+
+  1. Type  :!dir  or  :!ls  to get a listing of your directory.
+     You already know you must hit <ENTER> after this.
+
+  2. Choose a filename that does not exist yet, such as TEST.
+
+  3. Now type:	 :w TEST   (where TEST is the filename you chose.)
+
+  4. This saves the whole file (the Vim Tutor) under the name TEST.
+     To verify this, type    :!dir  or  :!ls   again to see your directory.
+
+NOTE: If you were to exit Vim and start it again with  vim TEST , the file
+      would be an exact copy of the tutor when you saved it.
+
+  5. Now remove the file by typing (Windows):   :!del TEST
+				or (Unix):	:!rm TEST
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		    Lesson 5.3: SELECTING TEXT TO WRITE
+
+
+	** To save part of the file, type  v  motion  :w FILENAME **
+
+  1. Move the cursor to this line.
+
+  2. Press  v  and move the cursor to the fifth item below.  Notice that the
+     text is highlighted.
+
+  3. Press the  :  character.  At the bottom of the screen  :'<,'> will appear.
+
+  4. Type  w TEST  , where TEST is a filename that does not exist yet.  Verify
+     that you see  :'<,'>w TEST  before you press <ENTER>.
+
+  5. Vim will write the selected lines to the file TEST.  Use  :!dir  or  :!ls
+     to see it.  Do not remove it yet!  We will use it in the next lesson.
+
+NOTE:  Pressing  v  starts Visual selection.  You can move the cursor around
+       to make the selection bigger or smaller.  Then you can use an operator
+       to do something with the text.  For example,  d  deletes the text.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		   Lesson 5.4: RETRIEVING AND MERGING FILES
+
+
+       ** To insert the contents of a file, type  :r FILENAME  **
+
+  1. Place the cursor just above this line.
+
+NOTE:  After executing Step 2 you will see text from lesson 5.3.  Then move
+       DOWN to see this lesson again.
+
+  2. Now retrieve your TEST file using the command   :r TEST   where TEST is
+     the name of the file you used.
+     The file you retrieve is placed below the cursor line.
+
+  3. To verify that a file was retrieved, cursor back and notice that there
+     are now two copies of lesson 5.3, the original and the file version.
+
+NOTE:  You can also read the output of an external command.  For example,
+       :r !ls  reads the output of the ls command and puts it below the
+       cursor.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 5 SUMMARY
+
+
+  1.  :!command  executes an external command.
+
+      Some useful examples are:
+	 (Windows)	  (Unix)
+	  :!dir		   :!ls		   -  shows a directory listing.
+	  :!del FILENAME   :!rm FILENAME   -  removes file FILENAME.
+
+  2.  :w FILENAME  writes the current Vim file to disk with name FILENAME.
+
+  3.  v  motion  :w FILENAME  saves the Visually selected lines in file
+      FILENAME.
+
+  4.  :r FILENAME  retrieves disk file FILENAME and puts it below the
+      cursor position.
+
+  5.  :r !dir  reads the output of the dir command and puts it below the
+      cursor position.
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			 Lesson 6.1: THE OPEN COMMAND
+
+
+ ** Type  o  to open a line below the cursor and place you in Insert mode. **
+
+  1. Move the cursor to the first line below marked --->.
+
+  2. Type the lowercase letter  o  to open up a line BELOW the cursor and place
+     you in Insert mode.
+
+  3. Now type some text and press <ESC> to exit Insert mode.
+
+---> After typing  o  the cursor is placed on the open line in Insert mode.
+
+  4. To open up a line ABOVE the cursor, simply type a capital	O , rather
+     than a lowercase  o.  Try this on the line below.
+
+---> Open up a line above this by typing O while the cursor is on this line.
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 6.2: THE APPEND COMMAND
+
+
+	     ** Type  a  to insert text AFTER the cursor. **
+
+  1. Move the cursor to the start of the first line below marked --->.
+
+  2. Press  e  until the cursor is on the end of  li .
+
+  3. Type an  a  (lowercase) to append text AFTER the cursor.
+
+  4. Complete the word like the line below it.  Press <ESC> to exit Insert
+     mode.
+
+  5. Use  e  to move to the next incomplete word and repeat steps 3 and 4.
+
+---> This li will allow you to pract appendi text to a line.
+---> This line will allow you to practice appending text to a line.
+
+NOTE:  a, i and A all go to the same Insert mode, the only difference is where
+       the characters are inserted.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		    Lesson 6.3: ANOTHER WAY TO REPLACE
+
+
+      ** Type a capital  R  to replace more than one character. **
+
+  1. Move the cursor to the first line below marked --->.  Move the cursor to
+     the beginning of the first  xxx .
+
+  2. Now press  R  and type the number below it in the second line, so that it
+     replaces the xxx .
+
+  3. Press <ESC> to leave Replace mode.  Notice that the rest of the line
+     remains unmodified.
+
+  4. Repeat the steps to replace the remaining xxx.
+
+---> Adding 123 to xxx gives you xxx.
+---> Adding 123 to 456 gives you 579.
+
+NOTE:  Replace mode is like Insert mode, but every typed character deletes an
+       existing character.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			Lesson 6.4: COPY AND PASTE TEXT
+
+
+	  ** Use the  y  operator to copy text and  p  to paste it **
+
+  1. Move to the line below marked ---> and place the cursor after "a)".
+
+  2. Start Visual mode with  v  and move the cursor to just before "first".
+
+  3. Type  y  to yank (copy) the highlighted text.
+
+  4. Move the cursor to the end of the next line:  j$
+
+  5. Type  p  to put (paste) the text.  Then type:  a second <ESC> .
+
+  6. Use Visual mode to select " item.", yank it with  y , move to the end of
+     the next line with  j$  and put the text there with  p .
+
+--->  a) this is the first item.
+      b)
+
+  NOTE: You can also use  y  as an operator;  yw  yanks one word.
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			    Lesson 6.5: SET OPTION
+
+
+	  ** Set an option so a search or substitute ignores case **
+
+  1. Search for 'ignore' by entering:  /ignore <ENTER>
+     Repeat several times by pressing  n .
+
+  2. Set the 'ic' (Ignore case) option by entering:   :set ic
+
+  3. Now search for 'ignore' again by pressing  n
+     Notice that Ignore and IGNORE are now also found.
+
+  4. Set the 'hlsearch' and 'incsearch' options:  :set hls is
+
+  5. Now type the search command again and see what happens:  /ignore <ENTER>
+
+  6. To disable ignoring case enter:  :set noic
+
+NOTE:  To remove the highlighting of matches enter:   :nohlsearch
+NOTE:  If you want to ignore case for just one search command, use  \c
+       in the phrase:  /ignore\c <ENTER>
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 6 SUMMARY
+
+  1. Type  o  to open a line BELOW the cursor and start Insert mode.
+     Type  O  to open a line ABOVE the cursor.
+
+  2. Type  a  to insert text AFTER the cursor.
+     Type  A  to insert text after the end of the line.
+
+  3. The  e  command moves to the end of a word.
+
+  4. The  y  operator yanks (copies) text,  p  puts (pastes) it.
+
+  5. Typing a capital  R  enters Replace mode until  <ESC>  is pressed.
+
+  6. Typing ":set xxx" sets the option "xxx".  Some options are:
+  	'ic' 'ignorecase'	ignore upper/lower case when searching
+	'is' 'incsearch'	show partial matches for a search phrase
+	'hls' 'hlsearch'	highlight all matching phrases
+     You can either use the long or the short option name.
+
+  7. Prepend "no" to switch an option off:   :set noic
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		       Lesson 7.1: GETTING HELP
+
+
+		      ** Use the on-line help system **
+
+  Vim has a comprehensive on-line help system.  To get started, try one of
+  these three:
+	- press the <HELP> key (if you have one)
+	- press the <F1> key (if you have one)
+	- type   :help <ENTER>
+
+  Read the text in the help window to find out how the help works.
+  Type  CTRL-W CTRL-W   to jump from one window to another.
+  Type    :q <ENTER>    to close the help window.
+
+  You can find help on just about any subject, by giving an argument to the
+  ":help" command.  Try these (don't forget pressing <ENTER>):
+
+	:help w
+	:help c_CTRL-D
+	:help insert-index
+	:help user-manual
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+		      Lesson 7.2: CREATE A STARTUP SCRIPT
+
+
+			  ** Enable Vim features **
+
+  Vim has many more features than Vi, but most of them are disabled by
+  default.  To start using more features you have to create a "vimrc" file.
+
+  1. Start editing the "vimrc" file.  This depends on your system:
+	:e ~/.vimrc		for Unix
+	:e $VIM/_vimrc		for Windows
+
+  2. Now read the example "vimrc" file contents:
+	:r $VIMRUNTIME/vimrc_example.vim
+
+  3. Write the file with:
+	:w
+
+  The next time you start Vim it will use syntax highlighting.
+  You can add all your preferred settings to this "vimrc" file.
+  For more information type  :help vimrc-intro
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			     Lesson 7.3: COMPLETION
+
+
+	      ** Command line completion with CTRL-D and <TAB> **
+
+  1. Make sure Vim is not in compatible mode:  :set nocp
+
+  2. Look what files exist in the directory:  :!ls   or  :!dir
+
+  3. Type the start of a command:  :e
+
+  4. Press  CTRL-D  and Vim will show a list of commands that start with "e".
+
+  5. Type  d<TAB>  and Vim will complete the command name to ":edit".
+
+  6. Now add a space and the start of an existing file name:  :edit FIL
+
+  7. Press <TAB>.  Vim will complete the name (if it is unique).
+
+NOTE:  Completion works for many commands.  Just try pressing CTRL-D and
+       <TAB>.  It is especially useful for  :help .
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+			       Lesson 7 SUMMARY
+
+
+  1. Type  :help  or press <F1> or <HELP>  to open a help window.
+
+  2. Type  :help cmd  to find help on  cmd .
+
+  3. Type  CTRL-W CTRL-W  to jump to another window.
+
+  4. Type  :q  to close the help window.
+
+  5. Create a vimrc startup script to keep your preferred settings.
+
+  6. When typing a  :  command, press CTRL-D to see possible completions.
+     Press <TAB> to use one completion.
+
+
+
+
+
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+  This concludes the Vim Tutor.  It was intended to give a brief overview of
+  the Vim editor, just enough to allow you to use the editor fairly easily.
+  It is far from complete as Vim has many many more commands.  Read the user
+  manual next: ":help user-manual".
+
+  For further reading and studying, this book is recommended:
+	Vim - Vi Improved - by Steve Oualline
+	Publisher: New Riders
+  The first book completely dedicated to Vim.  Especially useful for beginners.
+  There are many examples and pictures.
+  See http://iccf-holland.org/click5.html
+
+  This book is older and more about Vi than Vim, but also recommended:
+	Learning the Vi Editor - by Linda Lamb
+	Publisher: O'Reilly & Associates Inc.
+  It is a good book to get to know almost anything you want to do with Vi.
+  The sixth edition also includes information on Vim.
+
+  This tutorial was written by Michael C. Pierce and Robert K. Ware,
+  Colorado School of Mines using ideas supplied by Charles Smith,
+  Colorado State University.  E-mail: bware@mines.colorado.edu.
+
+  Modified for Vim by Bram Moolenaar.
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

.gitignore

@@ -0,0 +1,162 @@
+# User-defined
+# blahblah
+*.h5
+*.txt
+*.csv
+*.yaml
+config.py
+training_images/
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintainted in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+

Shoe Classifier_VGG16.ipynb

@@ -0,0 +1,398 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "572dc7fb",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2022-08-01 23:57:09.348119: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n"
+     ]
+    }
+   ],
+   "source": [
+    "from matplotlib import pyplot as plt\n",
+    "from matplotlib.image import imread\n",
+    "import pandas as pd\n",
+    "from collections import Counter\n",
+    "import json\n",
+    "import os\n",
+    "import re\n",
+    "import tempfile\n",
+    "import numpy as np\n",
+    "from os.path import exists\n",
+    "from imblearn.under_sampling import RandomUnderSampler\n",
+    "from PIL import ImageFile\n",
+    "import sklearn as sk\n",
+    "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n",
+    "import tensorflow as tf\n",
+    "import tensorflow.keras\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Flatten, Activation\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.optimizers import Adam\n",
+    "# custom modules\n",
+    "import image_faults\n",
+    "\n",
+    "ImageFile.LOAD_TRUNCATED_IMAGES = True"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "6ea418cc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def add_regularization(model, regularizer=tf.keras.regularizers.l2(0.0001)):\n",
+    "\n",
+    "    if not isinstance(regularizer, tf.keras.regularizers.Regularizer):\n",
+    "      print(\"Regularizer must be a subclass of tf.keras.regularizers.Regularizer\")\n",
+    "      return model\n",
+    "\n",
+    "    for layer in model.layers:\n",
+    "        for attr in ['kernel_regularizer']:\n",
+    "            if hasattr(layer, attr):\n",
+    "              setattr(layer, attr, regularizer)\n",
+    "\n",
+    "    # When we change the layers attributes, the change only happens in the model config file\n",
+    "    model_json = model.to_json()\n",
+    "\n",
+    "    # Save the weights before reloading the model.\n",
+    "    tmp_weights_path = os.path.join(tempfile.gettempdir(), 'tmp_weights.h5')\n",
+    "    model.save_weights(tmp_weights_path)\n",
+    "\n",
+    "    # load the model from the config\n",
+    "    model = tf.keras.models.model_from_json(model_json)\n",
+    "    \n",
+    "    # Reload the model weights\n",
+    "    model.load_weights(tmp_weights_path, by_name=True)\n",
+    "    return model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "a5c72863",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "image_faults.faulty_images() # removes faulty images\n",
+    "df = pd.read_csv('expanded_class.csv', index_col=[0], low_memory=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "1057a442",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "def dict_pics_jup():\n",
+    "    '''\n",
+    "    {source:target} dict used to replace source urls with image location as input\n",
+    "    '''\n",
+    "    target_dir = os.getcwd() + os.sep + \"training_images\"\n",
+    "    with open('temp_pics_source_list.txt') as f:\n",
+    "        temp_pics_source_list = json.load(f)\n",
+    "        \n",
+    "    dict_pics = {}\n",
+    "    for k in temp_pics_source_list:\n",
+    "         patt_1 = re.search(r'[^/]+(?=/\\$_|.(\\.jpg|\\.jpeg|\\.png))', k, re.IGNORECASE)\n",
+    "         patt_2 = re.search(r'(\\.jpg|\\.jpeg|\\.png)', k, re.IGNORECASE)\n",
+    "         if patt_1 and patt_2 is not None:\n",
+    "             tag = patt_1.group() + patt_2.group().lower()\n",
+    "             file_name = target_dir + os.sep + tag\n",
+    "             dict_pics.update({k:file_name})\n",
+    "    print(\"{source:target} dictionary created @ \" + target_dir)\n",
+    "    return dict_pics"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "7a6146e6",
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "expected string or bytes-like object",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-5-0009b269209e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdict_pics\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdict_pics_jup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'women_cat_list.txt'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m     \u001b[0mwomen_cats\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'men_cat_list.txt'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m<ipython-input-4-4701772f6383>\u001b[0m in \u001b[0;36mdict_pics_jup\u001b[0;34m()\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0mdict_pics\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtemp_pics_source_list\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m          \u001b[0mpatt_1\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mre\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msearch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mr'[^/]+(?=/\\$_|.(\\.jpg|\\.jpeg|\\.png))'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mre\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIGNORECASE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     12\u001b[0m          \u001b[0mpatt_2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mre\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msearch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mr'(\\.jpg|\\.jpeg|\\.png)'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mre\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIGNORECASE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     13\u001b[0m          \u001b[0;32mif\u001b[0m \u001b[0mpatt_1\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mpatt_2\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/lib/python3.8/re.py\u001b[0m in \u001b[0;36msearch\u001b[0;34m(pattern, string, flags)\u001b[0m\n\u001b[1;32m    199\u001b[0m     \"\"\"Scan through string looking for a match to the pattern, returning\n\u001b[1;32m    200\u001b[0m     a Match object, or None if no match was found.\"\"\"\n\u001b[0;32m--> 201\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0m_compile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpattern\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mflags\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msearch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstring\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    202\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    203\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0msub\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpattern\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrepl\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstring\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcount\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mflags\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mTypeError\u001b[0m: expected string or bytes-like object"
+     ]
+    }
+   ],
+   "source": [
+    "dict_pics = dict_pics_jup()\n",
+    "\n",
+    "with open('women_cat_list.txt') as f:\n",
+    "    women_cats = json.load(f)\n",
+    "with open('men_cat_list.txt') as f:\n",
+    "    men_cats = json.load(f)\n",
+    "    \n",
+    "with open('temp_pics_source_list.txt') as f:\n",
+    "    tempics = json.load(f)\n",
+    "# list of image urls that did not get named properly which will be removed from the dataframe\n",
+    "drop_row_vals = []\n",
+    "for pic in tempics:\n",
+    "    try:\n",
+    "        dict_pics[pic]\n",
+    "    except KeyError:\n",
+    "        drop_row_vals.append(pic)\n",
+    "\n",
+    "df['PrimaryCategoryID'] = df['PrimaryCategoryID'].astype(str) # pandas thinks ids are ints\n",
+    "ddf = df[df.PictureURL.isin(drop_row_vals)==False] # remove improperly named image files\n",
+    "df = ddf[ddf.PrimaryCategoryID.isin(men_cats)==False] # removes rows of womens categories\n",
+    "\n",
+    "blah = pd.Series(df.PictureURL)\n",
+    "df = df.drop(labels=['PictureURL'], axis=1)\n",
+    "\n",
+    "blah = blah.apply(lambda x: dict_pics[x])\n",
+    "df = pd.concat([blah, df],axis=1)\n",
+    "df = df.groupby('PrimaryCategoryID').filter(lambda x: len(x)>25) # removes cat outliers\n",
+    "\n",
+    "df=df.sample(frac=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8a3a86a1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "undersample = RandomUnderSampler(sampling_strategy='auto')\n",
+    "train, y_under = undersample.fit_resample(df, df['PrimaryCategoryID'])\n",
+    "print(Counter(train['PrimaryCategoryID']))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "506aa5cf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train, test = train_test_split(train, test_size=0.2, random_state=42)\n",
+    "# stratify=train['PrimaryCategoryID']\n",
+    "# train['PrimaryCategoryID'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "4d72eb90",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "datagen = ImageDataGenerator(rescale=1./255., \n",
+    "                             validation_split=.2,\n",
+    "                             #samplewise_std_normalization=True,\n",
+    "                             #horizontal_flip= True,\n",
+    "                             #vertical_flip= True,\n",
+    "                             #width_shift_range= 0.2,\n",
+    "                             #height_shift_range= 0.2,\n",
+    "                             #rotation_range= 90,\n",
+    "                             preprocessing_function=tf.keras.applications.vgg16.preprocess_input)\n",
+    "train_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)],\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=32,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(244,244),\n",
+    "    subset='training'\n",
+    "    )\n",
+    "validation_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)], # is using train right?\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=32,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(244,244),\n",
+    "    subset='validation'\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7b70f37f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "imgs, labels = next(train_generator)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1ed54bf5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plotImages(images_arr):\n",
+    "    fig, axes = plt.subplots(1, 10, figsize=(20,20))\n",
+    "    axes = axes.flatten()\n",
+    "    for img, ax in zip( images_arr, axes):\n",
+    "        ax.imshow(img)\n",
+    "        ax.axis('off')\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "85934565",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#plotImages(imgs)\n",
+    "#print(labels)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6322bcad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "physical_devices = tf.config.list_physical_devices('GPU')\n",
+    "print(len(physical_devices))\n",
+    "tf.config.experimental.set_memory_growth(physical_devices[0], True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b31af79e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "base_model = tf.keras.applications.vgg16.VGG16(weights='imagenet')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fe06f2bf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#model = Sequential()\n",
+    "#for layer in base_model.layers[:-1]:\n",
+    "#    model.add(layer)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7d3cc82c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# loop through layers, add Dropout after layers 'fc1' and 'fc2'\n",
+    "updated_model = Sequential()\n",
+    "for layer in base_model.layers[:-1]:\n",
+    "    updated_model.add(layer)\n",
+    "    if layer.name in ['fc1', 'fc2']:\n",
+    "        updated_model.add(Dropout(.50))\n",
+    "\n",
+    "model = updated_model\n",
+    "\n",
+    "for layer in model.layers:\n",
+    "    layer.trainable = True\n",
+    "\n",
+    "model.add(Dense(units=7, activation='softmax'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c774d787",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "#model = add_regularization(model)\n",
+    "model.summary()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fd5d1246",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model.compile(optimizer=Adam(learning_rate=.0001), loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9cd2ba27",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [],
+   "source": [
+    "model.fit(x=train_generator,\n",
+    "          steps_per_epoch=len(train_generator),\n",
+    "          validation_data=validation_generator,\n",
+    "          validation_steps=len(validation_generator),\n",
+    "          epochs=30,\n",
+    "          verbose=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "63f791af",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

Shoe Classifier_VGG19.ipynb

@@ -0,0 +1,522 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "572dc7fb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from matplotlib import pyplot as plt\n",
+    "from matplotlib.image import imread\n",
+    "import pandas as pd\n",
+    "from collections import Counter\n",
+    "import json\n",
+    "import os\n",
+    "import re\n",
+    "import tempfile\n",
+    "import numpy as np\n",
+    "from os.path import exists\n",
+    "from imblearn.under_sampling import RandomUnderSampler\n",
+    "from PIL import ImageFile\n",
+    "import sklearn as sk\n",
+    "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n",
+    "import tensorflow as tf\n",
+    "import tensorflow.keras\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Flatten, Activation\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.optimizers import Adam\n",
+    "# custom modules\n",
+    "import image_faults\n",
+    "\n",
+    "ImageFile.LOAD_TRUNCATED_IMAGES = True"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def add_regularization(model, regularizer=tf.keras.regularizers.l2(0.0001)):\n",
+    "\n",
+    "    if not isinstance(regularizer, tf.keras.regularizers.Regularizer):\n",
+    "      print(\"Regularizer must be a subclass of tf.keras.regularizers.Regularizer\")\n",
+    "      return model\n",
+    "\n",
+    "    for layer in model.layers:\n",
+    "        for attr in ['kernel_regularizer']:\n",
+    "            if hasattr(layer, attr):\n",
+    "              setattr(layer, attr, regularizer)\n",
+    "\n",
+    "    # When we change the layers attributes, the change only happens in the model config file\n",
+    "    model_json = model.to_json()\n",
+    "\n",
+    "    # Save the weights before reloading the model.\n",
+    "    tmp_weights_path = os.path.join(tempfile.gettempdir(), 'tmp_weights.h5')\n",
+    "    model.save_weights(tmp_weights_path)\n",
+    "\n",
+    "    # load the model from the config\n",
+    "    model = tf.keras.models.model_from_json(model_json)\n",
+    "    \n",
+    "    # Reload the model weights\n",
+    "    model.load_weights(tmp_weights_path, by_name=True)\n",
+    "    return model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "a5c72863",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# image_faults.faulty_images() # removes faulty images\n",
+    "df = pd.read_csv('expanded_class.csv', index_col=[0], low_memory=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "1057a442",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{source:target} dictionary created @ /tf/training_images\n"
+     ]
+    }
+   ],
+   "source": [
+    "def dict_pics_jup():\n",
+    "    '''\n",
+    "    {source:target} dict used to replace source urls with image location as input\n",
+    "    '''\n",
+    "    target_dir = os.getcwd() + os.sep + \"training_images\"\n",
+    "    with open('temp_pics_source_list.txt') as f:\n",
+    "        temp_pics_source_list = json.load(f)\n",
+    "        \n",
+    "    dict_pics = {}\n",
+    "    for k in temp_pics_source_list:\n",
+    "         patt_1 = re.search(r'[^/]+(?=/\\$_|.(\\.jpg|\\.jpeg|\\.png))', k, re.IGNORECASE)\n",
+    "         patt_2 = re.search(r'(\\.jpg|\\.jpeg|\\.png)', k, re.IGNORECASE)\n",
+    "         if patt_1 and patt_2 is not None:\n",
+    "             tag = patt_1.group() + patt_2.group().lower()\n",
+    "             file_name = target_dir + os.sep + tag\n",
+    "             dict_pics.update({k:file_name})\n",
+    "    print(\"{source:target} dictionary created @ \" + target_dir)\n",
+    "    return dict_pics"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "7a6146e6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dict_pics = dict_pics_jup()\n",
+    "\n",
+    "with open('women_cat_list.txt') as f:\n",
+    "    women_cats = json.load(f)\n",
+    "with open('men_cat_list.txt') as f:\n",
+    "    men_cats = json.load(f)\n",
+    "    \n",
+    "with open('temp_pics_source_list.txt') as f:\n",
+    "    tempics = json.load(f)\n",
+    "# list of image urls that did not get named properly which will be removed from the dataframe\n",
+    "drop_row_vals = []\n",
+    "for pic in tempics:\n",
+    "    try:\n",
+    "        dict_pics[pic]\n",
+    "    except KeyError:\n",
+    "        drop_row_vals.append(pic)\n",
+    "\n",
+    "df['PrimaryCategoryID'] = df['PrimaryCategoryID'].astype(str) # pandas thinks ids are ints\n",
+    "ddf = df[df.PictureURL.isin(drop_row_vals)==False] # remove improperly named image files\n",
+    "df = ddf[ddf.PrimaryCategoryID.isin(men_cats)==False] # removes rows of womens categories\n",
+    "\n",
+    "blah = pd.Series(df.PictureURL)\n",
+    "df = df.drop(labels=['PictureURL'], axis=1)\n",
+    "\n",
+    "blah = blah.apply(lambda x: dict_pics[x])\n",
+    "df = pd.concat([blah, df],axis=1)\n",
+    "df = df.groupby('PrimaryCategoryID').filter(lambda x: len(x)>25) # removes cat outliers\n",
+    "\n",
+    "df=df.sample(frac=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Counter({'11498': 3913, '11504': 3913, '11505': 3913, '11632': 3913, '15709': 3913, '24087': 3913, '45333': 3913, '53120': 3913, '53548': 3913, '53557': 3913, '55793': 3913, '62107': 3913, '95672': 3913})\n"
+     ]
+    }
+   ],
+   "source": [
+    "undersample = RandomUnderSampler(sampling_strategy='auto')\n",
+    "train, y_under = undersample.fit_resample(df, df['PrimaryCategoryID'])\n",
+    "#print(Counter(train['PrimaryCategoryID']))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "506aa5cf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train, test = train_test_split(train, test_size=0.2, random_state=42)\n",
+    "# stratify=train['PrimaryCategoryID']\n",
+    "# train['PrimaryCategoryID'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "4d72eb90",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 32547 validated image filenames belonging to 13 classes.\n",
+      "Found 8136 validated image filenames belonging to 13 classes.\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/usr/local/lib/python3.8/dist-packages/keras_preprocessing/image/dataframe_iterator.py:279: UserWarning: Found 12 invalid image filename(s) in x_col=\"PictureURL\". These filename(s) will be ignored.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
+   "source": [
+    "datagen = ImageDataGenerator(rescale=1./255., \n",
+    "                             validation_split=.2,\n",
+    "                             #samplewise_std_normalization=True,\n",
+    "                             #horizontal_flip= True,\n",
+    "                             #vertical_flip= True,\n",
+    "                             #width_shift_range= 0.2,\n",
+    "                             #height_shift_range= 0.2,\n",
+    "                             #rotation_range= 90,\n",
+    "                             preprocessing_function=tf.keras.applications.vgg16.preprocess_input)\n",
+    "train_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)],\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=32,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(244,244),\n",
+    "    subset='training'\n",
+    "    )\n",
+    "validation_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)], # is using train right?\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=32,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(244,244),\n",
+    "    subset='validation'\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "7b70f37f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "imgs, labels = next(train_generator)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "1ed54bf5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plotImages(images_arr):\n",
+    "    fig, axes = plt.subplots(1, 10, figsize=(20,20))\n",
+    "    axes = axes.flatten()\n",
+    "    for img, ax in zip( images_arr, axes):\n",
+    "        ax.imshow(img)\n",
+    "        ax.axis('off')\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "85934565",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#plotImages(imgs)\n",
+    "#print(labels)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "6322bcad",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1\n"
+     ]
+    }
+   ],
+   "source": [
+    "physical_devices = tf.config.list_physical_devices('GPU')\n",
+    "print(len(physical_devices))\n",
+    "tf.config.experimental.set_memory_growth(physical_devices[0], True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "b31af79e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "base_model = tf.keras.applications.vgg19.VGG19(weights='imagenet')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "fe06f2bf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model = Sequential()\n",
+    "for layer in base_model.layers[:-1]:\n",
+    "    model.add(layer)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "7d3cc82c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for layer in model.layers:\n",
+    "    layer.trainable = True\n",
+    "    \n",
+    "#model.add(Dropout(.5))\n",
+    "#model.add(Dense(64, activation='softmax'))\n",
+    "# model.add(Dropout(.25))\n",
+    "\n",
+    "model.add(Dense(units=7, activation='softmax'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "c774d787",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      \n",
+      "                                                                 \n",
+      " block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     \n",
+      "                                                                 \n",
+      " block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         \n",
+      "                                                                 \n",
+      " block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     \n",
+      "                                                                 \n",
+      " block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    \n",
+      "                                                                 \n",
+      " block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         \n",
+      "                                                                 \n",
+      " block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    \n",
+      "                                                                 \n",
+      " block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    \n",
+      "                                                                 \n",
+      " block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    \n",
+      "                                                                 \n",
+      " block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         \n",
+      "                                                                 \n",
+      " block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   \n",
+      "                                                                 \n",
+      " block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
+      "                                                                 \n",
+      " block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
+      "                                                                 \n",
+      " block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         \n",
+      "                                                                 \n",
+      " block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
+      "                                                                 \n",
+      " block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
+      "                                                                 \n",
+      " block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
+      "                                                                 \n",
+      " block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         \n",
+      "                                                                 \n",
+      " flatten (Flatten)           (None, 25088)             0         \n",
+      "                                                                 \n",
+      " fc1 (Dense)                 (None, 4096)              102764544 \n",
+      "                                                                 \n",
+      " fc2 (Dense)                 (None, 4096)              16781312  \n",
+      "                                                                 \n",
+      " dense (Dense)               (None, 13)                53261     \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 134,313,805\n",
+      "Trainable params: 134,313,805\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "#model = add_regularization(model)\n",
+    "model.summary()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "fd5d1246",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model.compile(optimizer=Adam(learning_rate=.0001), loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])\n",
+    "# sparse_categorical_crossentropy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "9cd2ba27",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/30\n",
+      "1018/1018 [==============================] - 391s 379ms/step - loss: 2.4329 - accuracy: 0.1571 - val_loss: 2.1902 - val_accuracy: 0.2525\n",
+      "Epoch 2/30\n",
+      "1018/1018 [==============================] - 379s 373ms/step - loss: 2.0578 - accuracy: 0.2960 - val_loss: 1.9160 - val_accuracy: 0.3330\n",
+      "Epoch 3/30\n",
+      "1018/1018 [==============================] - 381s 374ms/step - loss: 1.8221 - accuracy: 0.3681 - val_loss: 1.8588 - val_accuracy: 0.3535\n",
+      "Epoch 4/30\n",
+      "1018/1018 [==============================] - 383s 376ms/step - loss: 1.6406 - accuracy: 0.4272 - val_loss: 1.7819 - val_accuracy: 0.4028\n",
+      "Epoch 5/30\n",
+      "1018/1018 [==============================] - 383s 376ms/step - loss: 1.4577 - accuracy: 0.4920 - val_loss: 1.7216 - val_accuracy: 0.4158\n",
+      "Epoch 6/30\n",
+      "1018/1018 [==============================] - 379s 372ms/step - loss: 1.2528 - accuracy: 0.5607 - val_loss: 1.7924 - val_accuracy: 0.4140\n",
+      "Epoch 7/30\n",
+      "1018/1018 [==============================] - 378s 371ms/step - loss: 1.0030 - accuracy: 0.6469 - val_loss: 1.8017 - val_accuracy: 0.4303\n",
+      "Epoch 8/30\n",
+      "1018/1018 [==============================] - 379s 372ms/step - loss: 0.7405 - accuracy: 0.7420 - val_loss: 1.9863 - val_accuracy: 0.4453\n",
+      "Epoch 9/30\n",
+      "1018/1018 [==============================] - 379s 372ms/step - loss: 0.4704 - accuracy: 0.8354 - val_loss: 2.3988 - val_accuracy: 0.4263\n",
+      "Epoch 10/30\n",
+      "1018/1018 [==============================] - 379s 372ms/step - loss: 0.3059 - accuracy: 0.8944 - val_loss: 2.7526 - val_accuracy: 0.4303\n",
+      "Epoch 11/30\n",
+      "1018/1018 [==============================] - 377s 371ms/step - loss: 0.2160 - accuracy: 0.9278 - val_loss: 3.0618 - val_accuracy: 0.4250\n",
+      "Epoch 12/30\n",
+      " 437/1018 [===========>..................] - ETA: 2:53 - loss: 0.1370 - accuracy: 0.9536"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-20-4cd4443bbf2a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m model.fit(x=train_generator,\n\u001b[0m\u001b[1;32m      2\u001b[0m           \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_generator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m           \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvalidation_generator\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m           \u001b[0mvalidation_steps\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvalidation_generator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m           \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m30\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/keras/utils/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     62\u001b[0m     \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     63\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 64\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     65\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1214\u001b[0m                 _r=1):\n\u001b[1;32m   1215\u001b[0m               \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1216\u001b[0;31m               \u001b[0mtmp_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1217\u001b[0m               \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1218\u001b[0m                 \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/util/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    148\u001b[0m     \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    149\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 150\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    151\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    152\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    908\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    909\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mOptionalXlaContext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jit_compile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 910\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    911\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    912\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexperimental_get_tracing_count\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    940\u001b[0m       \u001b[0;31m# In this case we have created variables on the first call, so we run the\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    941\u001b[0m       \u001b[0;31m# defunned version which is guaranteed to never create variables.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 942\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateless_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# pylint: disable=not-callable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    943\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateful_fn\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    944\u001b[0m       \u001b[0;31m# Release the lock early so that multiple threads can perform the call\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   3128\u001b[0m       (graph_function,\n\u001b[1;32m   3129\u001b[0m        filtered_flat_args) = self._maybe_define_function(args, kwargs)\n\u001b[0;32m-> 3130\u001b[0;31m     return graph_function._call_flat(\n\u001b[0m\u001b[1;32m   3131\u001b[0m         filtered_flat_args, captured_inputs=graph_function.captured_inputs)  # pylint: disable=protected-access\n\u001b[1;32m   3132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_call_flat\u001b[0;34m(self, args, captured_inputs, cancellation_manager)\u001b[0m\n\u001b[1;32m   1957\u001b[0m         and executing_eagerly):\n\u001b[1;32m   1958\u001b[0m       \u001b[0;31m# No tape is watching; skip to running the function.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1959\u001b[0;31m       return self._build_call_outputs(self._inference_function.call(\n\u001b[0m\u001b[1;32m   1960\u001b[0m           ctx, args, cancellation_manager=cancellation_manager))\n\u001b[1;32m   1961\u001b[0m     forward_backward = self._select_forward_and_backward_functions(\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, ctx, args, cancellation_manager)\u001b[0m\n\u001b[1;32m    596\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0m_InterpolateFunctionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    597\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcancellation_manager\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 598\u001b[0;31m           outputs = execute.execute(\n\u001b[0m\u001b[1;32m    599\u001b[0m               \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msignature\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    600\u001b[0m               \u001b[0mnum_outputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_outputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.8/dist-packages/tensorflow/python/eager/execute.py\u001b[0m in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     56\u001b[0m   \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     57\u001b[0m     \u001b[0mctx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mensure_initialized\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 58\u001b[0;31m     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,\n\u001b[0m\u001b[1;32m     59\u001b[0m                                         inputs, attrs, num_outputs)\n\u001b[1;32m     60\u001b[0m   \u001b[0;32mexcept\u001b[0m \u001b[0mcore\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_NotOkStatusException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "model.fit(x=train_generator,\n",
+    "          steps_per_epoch=len(train_generator),\n",
+    "          validation_data=validation_generator,\n",
+    "          validation_steps=len(validation_generator),\n",
+    "          epochs=30,\n",
+    "          verbose=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "63f791af",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

Shoe Classifier_Xception.ipynb

@@ -0,0 +1,882 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "572dc7fb",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2022-08-01 22:09:35.958273: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n"
+     ]
+    }
+   ],
+   "source": [
+    "from matplotlib import pyplot as plt\n",
+    "import cv\n",
+    "from matplotlib.image import imread\n",
+    "import pandas as pd\n",
+    "from collections import Counter\n",
+    "import json\n",
+    "import os\n",
+    "import re\n",
+    "import tempfile\n",
+    "import numpy as np\n",
+    "from os.path import exists\n",
+    "from imblearn.under_sampling import RandomUnderSampler\n",
+    "from PIL import ImageFile\n",
+    "import sklearn as sk\n",
+    "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n",
+    "import tensorflow as tf\n",
+    "import tensorflow.keras\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Flatten, Activation\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.optimizers import Adam\n",
+    "# custom modules\n",
+    "import image_faults\n",
+    "\n",
+    "ImageFile.LOAD_TRUNCATED_IMAGES = True"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "a5c72863",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "image_faults.faulty_images() # removes faulty images\n",
+    "df = pd.read_csv('expanded_class.csv', index_col=[0], low_memory=False)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "67ecdebe",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1')\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2022-08-01 22:09:39.570503: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set\n",
+      "2022-08-01 22:09:39.571048: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1\n",
+      "2022-08-01 22:09:39.613420: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.613584: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: \n",
+      "pciBusID: 0000:04:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6\n",
+      "coreClock: 1.725GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s\n",
+      "2022-08-01 22:09:39.613631: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.613751: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 1 with properties: \n",
+      "pciBusID: 0000:0b:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6\n",
+      "coreClock: 1.8GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s\n",
+      "2022-08-01 22:09:39.613767: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
+      "2022-08-01 22:09:39.614548: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10\n",
+      "2022-08-01 22:09:39.614572: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10\n",
+      "2022-08-01 22:09:39.615415: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10\n",
+      "2022-08-01 22:09:39.615547: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10\n",
+      "2022-08-01 22:09:39.616317: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10\n",
+      "2022-08-01 22:09:39.616763: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10\n",
+      "2022-08-01 22:09:39.618472: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7\n",
+      "2022-08-01 22:09:39.618532: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.618687: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.618830: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.618969: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.619075: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0, 1\n",
+      "2022-08-01 22:09:39.619877: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
+      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "2022-08-01 22:09:39.621856: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set\n",
+      "2022-08-01 22:09:39.792333: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.792467: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: \n",
+      "pciBusID: 0000:04:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6\n",
+      "coreClock: 1.725GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s\n",
+      "2022-08-01 22:09:39.792551: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.792644: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 1 with properties: \n",
+      "pciBusID: 0000:0b:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6\n",
+      "coreClock: 1.8GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s\n",
+      "2022-08-01 22:09:39.792680: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
+      "2022-08-01 22:09:39.792696: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10\n",
+      "2022-08-01 22:09:39.792706: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10\n",
+      "2022-08-01 22:09:39.792715: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10\n",
+      "2022-08-01 22:09:39.792724: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10\n",
+      "2022-08-01 22:09:39.792733: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10\n",
+      "2022-08-01 22:09:39.792741: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10\n",
+      "2022-08-01 22:09:39.792750: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7\n",
+      "2022-08-01 22:09:39.792797: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.792931: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.793053: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.793172: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:39.793263: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0, 1\n",
+      "2022-08-01 22:09:39.793290: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
+      "2022-08-01 22:09:41.188032: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:\n",
+      "2022-08-01 22:09:41.188052: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0 1 \n",
+      "2022-08-01 22:09:41.188057: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N N \n",
+      "2022-08-01 22:09:41.188059: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 1:   N N \n",
+      "2022-08-01 22:09:41.188316: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.188469: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.188599: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.188726: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.188831: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 22425 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:04:00.0, compute capability: 8.6)\n",
+      "2022-08-01 22:09:41.189525: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.189665: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-08-01 22:09:41.189758: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 21683 MB memory) -> physical GPU (device: 1, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:0b:00.0, compute capability: 8.6)\n"
+     ]
+    }
+   ],
+   "source": [
+    "mirrored_strategy = tf.distribute.MirroredStrategy(devices=[\"/gpu:0\",\"/gpu:1\"])\n",
+    "#\"/gpu:0\","
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "a89913e0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def dict_pics_jup():\n",
+    "    '''\n",
+    "    {source:target} dict used to replace source urls with image location as input\n",
+    "    '''\n",
+    "    target_dir = os.getcwd() + os.sep + \"training_images\"\n",
+    "    with open('temp_pics_source_list.txt') as f:\n",
+    "        temp_pics_source_list = json.load(f)\n",
+    "        \n",
+    "    dict_pics = {}\n",
+    "    for k in temp_pics_source_list:\n",
+    "        try:      \n",
+    "             patt_1 = re.search(r'[^/]+(?=/\\$_|.(\\.jpg|\\.jpeg|\\.png))', k, re.IGNORECASE)\n",
+    "             patt_2 = re.search(r'(\\.jpg|\\.jpeg|\\.png)', k, re.IGNORECASE)\n",
+    "             if patt_1 and patt_2 is not None:\n",
+    "                 tag = patt_1.group() + patt_2.group().lower()\n",
+    "                 file_name = target_dir + os.sep + tag\n",
+    "                 dict_pics.update({k:file_name})\n",
+    "        except TypeError:\n",
+    "            print(k)\n",
+    "    print(\"{source:target} dictionary created @ \" + target_dir)\n",
+    "    return dict_pics\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "id": "1057a442",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "nan\n",
+      "{source:target} dictionary created @ /home/unknown/Sync/projects/ebay ML Lister Project/training_images\n"
+     ]
+    }
+   ],
+   "source": [
+    "dict_pics = dict_pics_jup()\n",
+    "\n",
+    "with open('women_cat_list.txt') as f:\n",
+    "    women_cats = json.load(f)\n",
+    "with open('men_cat_list.txt') as f:\n",
+    "    men_cats = json.load(f)\n",
+    "    \n",
+    "with open('temp_pics_source_list.txt') as f:\n",
+    "    tempics = json.load(f)\n",
+    "# list of image urls that did not get named properly which will be removed from the dataframe\n",
+    "drop_row_vals = []\n",
+    "for pic in tempics:\n",
+    "    try:\n",
+    "        dict_pics[pic]\n",
+    "    except KeyError:\n",
+    "        drop_row_vals.append(pic)\n",
+    "\n",
+    "df['PrimaryCategoryID'] = df['PrimaryCategoryID'].astype(str) # pandas thinks ids are ints\n",
+    "df = df[df.PictureURL.isin(drop_row_vals)==False] # remove improperly named image files\n",
+    "df = df[df.PrimaryCategoryID.isin(men_cats)==False] # removes rows of womens categories\n",
+    "\n",
+    "blah = pd.Series(df.PictureURL)\n",
+    "df = df.drop(labels=['PictureURL'], axis=1)\n",
+    "\n",
+    "blah = blah.apply(lambda x: dict_pics[x])\n",
+    "df = pd.concat([blah, df],axis=1)\n",
+    "df = df.groupby('PrimaryCategoryID').filter(lambda x: len(x)>25) # removes cat outliers"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "id": "7a6146e6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'/home/unknown/Sync/projects/ebay ML Lister Project/training_images/7BQAAOSw0eZhpmqM.jpg'"
+      ]
+     },
+     "execution_count": 78,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df=df.sample(frac=1)\n",
+    "something = df.iloc[1,0]\n",
+    "something"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "id": "114cc3c0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "undersample = RandomUnderSampler(sampling_strategy='auto')\n",
+    "train, y_under = undersample.fit_resample(df, df['PrimaryCategoryID'])\n",
+    "#print(Counter(train['PrimaryCategoryID']))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "id": "506aa5cf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train, test = train_test_split(train, test_size=0.2, random_state=42)\n",
+    "# stratify=train['PrimaryCategoryID']\n",
+    "# train['PrimaryCategoryID'].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 80,
+   "id": "4d72eb90",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/unknown/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/keras_preprocessing/image/dataframe_iterator.py:279: UserWarning: Found 5 invalid image filename(s) in x_col=\"PictureURL\". These filename(s) will be ignored.\n",
+      "  warnings.warn(\n",
+      "/home/unknown/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/keras_preprocessing/image/dataframe_iterator.py:279: UserWarning: Found 5 invalid image filename(s) in x_col=\"PictureURL\". These filename(s) will be ignored.\n",
+      "  warnings.warn(\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 43744 validated image filenames belonging to 7 classes.\n",
+      "Found 10935 validated image filenames belonging to 7 classes.\n"
+     ]
+    }
+   ],
+   "source": [
+    "datagen = ImageDataGenerator(rescale=1./255., \n",
+    "                             validation_split=.2,\n",
+    "                             #samplewise_std_normalization=True,\n",
+    "                             #horizontal_flip= True,\n",
+    "                             #vertical_flip= True,\n",
+    "                             #width_shift_range= 0.2,\n",
+    "                             #height_shift_range= 0.2,\n",
+    "                             #rotation_range= 90,\n",
+    "                             preprocessing_function=tf.keras.applications.xception.preprocess_input)\n",
+    "\n",
+    "train_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)],\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=56,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(299,299),\n",
+    "    subset='training'\n",
+    "    )\n",
+    "validation_generator=datagen.flow_from_dataframe(\n",
+    "    dataframe=train[:len(train)], # is using train right?\n",
+    "    directory='./training_images',\n",
+    "    x_col='PictureURL',\n",
+    "    y_col='PrimaryCategoryID',\n",
+    "    batch_size=56,\n",
+    "    seed=42,\n",
+    "    shuffle=True,\n",
+    "    target_size=(299,299),\n",
+    "    subset='validation'\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 81,
+   "id": "7b70f37f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "imgs, labels = next(train_generator)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 82,
+   "id": "1ed54bf5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plotImages(images_arr):\n",
+    "    fig, axes = plt.subplots(1, 10, figsize=(20,20))\n",
+    "    axes = axes.flatten()\n",
+    "    for img, ax in zip( images_arr, axes):\n",
+    "        ax.imshow(img)\n",
+    "        ax.axis('off')\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 83,
+   "id": "85934565",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
+      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1440x1440 with 10 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plotImages(imgs)\n",
+    "# image = plt.imread('training_images/0t0AAOSw4tNgSQ1j.jpg')\n",
+    "# plt.imshow(image)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 84,
+   "id": "6322bcad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#physical_devices = tf.config.list_physical_devices('GPU')\n",
+    "#print(len(physical_devices))\n",
+    "#print(physical_devices)\n",
+    "#for gpu_instance in physical_devices: \n",
+    "#    tf.config.experimental.set_memory_growth(gpu_instance, True)\n",
+    "#tf.config.experimental.set_memory_growth(physical_devices[0], True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 85,
+   "id": "07fd25c6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# see https://www.kaggle.com/dmitrypukhov/cnn-with-imagedatagenerator-flow-from-dataframe for train/test/val split \n",
+    "# example\n",
+    "\n",
+    "# may need to either create a test dataset from the original dataset or just download a new one"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 86,
+   "id": "fe06f2bf",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"model_5\"\n",
+      "__________________________________________________________________________________________________\n",
+      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
+      "==================================================================================================\n",
+      "input_6 (InputLayer)            [(None, 299, 299, 3) 0                                            \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv1 (Conv2D)           (None, 149, 149, 32) 864         input_6[0][0]                    \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv1_bn (BatchNormaliza (None, 149, 149, 32) 128         block1_conv1[0][0]               \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv1_act (Activation)   (None, 149, 149, 32) 0           block1_conv1_bn[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv2 (Conv2D)           (None, 147, 147, 64) 18432       block1_conv1_act[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv2_bn (BatchNormaliza (None, 147, 147, 64) 256         block1_conv2[0][0]               \n",
+      "__________________________________________________________________________________________________\n",
+      "block1_conv2_act (Activation)   (None, 147, 147, 64) 0           block1_conv2_bn[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_sepconv1 (SeparableConv2 (None, 147, 147, 128 8768        block1_conv2_act[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_sepconv1_bn (BatchNormal (None, 147, 147, 128 512         block2_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_sepconv2_act (Activation (None, 147, 147, 128 0           block2_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_sepconv2 (SeparableConv2 (None, 147, 147, 128 17536       block2_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_sepconv2_bn (BatchNormal (None, 147, 147, 128 512         block2_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "conv2d_20 (Conv2D)              (None, 74, 74, 128)  8192        block1_conv2_act[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block2_pool (MaxPooling2D)      (None, 74, 74, 128)  0           block2_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "batch_normalization_20 (BatchNo (None, 74, 74, 128)  512         conv2d_20[0][0]                  \n",
+      "__________________________________________________________________________________________________\n",
+      "add_60 (Add)                    (None, 74, 74, 128)  0           block2_pool[0][0]                \n",
+      "                                                                 batch_normalization_20[0][0]     \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv1_act (Activation (None, 74, 74, 128)  0           add_60[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv1 (SeparableConv2 (None, 74, 74, 256)  33920       block3_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv1_bn (BatchNormal (None, 74, 74, 256)  1024        block3_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv2_act (Activation (None, 74, 74, 256)  0           block3_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv2 (SeparableConv2 (None, 74, 74, 256)  67840       block3_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_sepconv2_bn (BatchNormal (None, 74, 74, 256)  1024        block3_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "conv2d_21 (Conv2D)              (None, 37, 37, 256)  32768       add_60[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block3_pool (MaxPooling2D)      (None, 37, 37, 256)  0           block3_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "batch_normalization_21 (BatchNo (None, 37, 37, 256)  1024        conv2d_21[0][0]                  \n",
+      "__________________________________________________________________________________________________\n",
+      "add_61 (Add)                    (None, 37, 37, 256)  0           block3_pool[0][0]                \n",
+      "                                                                 batch_normalization_21[0][0]     \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv1_act (Activation (None, 37, 37, 256)  0           add_61[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv1 (SeparableConv2 (None, 37, 37, 728)  188672      block4_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv1_bn (BatchNormal (None, 37, 37, 728)  2912        block4_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv2_act (Activation (None, 37, 37, 728)  0           block4_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv2 (SeparableConv2 (None, 37, 37, 728)  536536      block4_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_sepconv2_bn (BatchNormal (None, 37, 37, 728)  2912        block4_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "conv2d_22 (Conv2D)              (None, 19, 19, 728)  186368      add_61[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block4_pool (MaxPooling2D)      (None, 19, 19, 728)  0           block4_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "batch_normalization_22 (BatchNo (None, 19, 19, 728)  2912        conv2d_22[0][0]                  \n",
+      "__________________________________________________________________________________________________\n",
+      "add_62 (Add)                    (None, 19, 19, 728)  0           block4_pool[0][0]                \n",
+      "                                                                 batch_normalization_22[0][0]     \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv1_act (Activation (None, 19, 19, 728)  0           add_62[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv1 (SeparableConv2 (None, 19, 19, 728)  536536      block5_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv1_bn (BatchNormal (None, 19, 19, 728)  2912        block5_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv2_act (Activation (None, 19, 19, 728)  0           block5_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv2 (SeparableConv2 (None, 19, 19, 728)  536536      block5_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv2_bn (BatchNormal (None, 19, 19, 728)  2912        block5_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv3_act (Activation (None, 19, 19, 728)  0           block5_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv3 (SeparableConv2 (None, 19, 19, 728)  536536      block5_sepconv3_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block5_sepconv3_bn (BatchNormal (None, 19, 19, 728)  2912        block5_sepconv3[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "add_63 (Add)                    (None, 19, 19, 728)  0           block5_sepconv3_bn[0][0]         \n",
+      "                                                                 add_62[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv1_act (Activation (None, 19, 19, 728)  0           add_63[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv1 (SeparableConv2 (None, 19, 19, 728)  536536      block6_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv1_bn (BatchNormal (None, 19, 19, 728)  2912        block6_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv2_act (Activation (None, 19, 19, 728)  0           block6_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv2 (SeparableConv2 (None, 19, 19, 728)  536536      block6_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv2_bn (BatchNormal (None, 19, 19, 728)  2912        block6_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv3_act (Activation (None, 19, 19, 728)  0           block6_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv3 (SeparableConv2 (None, 19, 19, 728)  536536      block6_sepconv3_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block6_sepconv3_bn (BatchNormal (None, 19, 19, 728)  2912        block6_sepconv3[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "add_64 (Add)                    (None, 19, 19, 728)  0           block6_sepconv3_bn[0][0]         \n",
+      "                                                                 add_63[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv1_act (Activation (None, 19, 19, 728)  0           add_64[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv1 (SeparableConv2 (None, 19, 19, 728)  536536      block7_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv1_bn (BatchNormal (None, 19, 19, 728)  2912        block7_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv2_act (Activation (None, 19, 19, 728)  0           block7_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv2 (SeparableConv2 (None, 19, 19, 728)  536536      block7_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv2_bn (BatchNormal (None, 19, 19, 728)  2912        block7_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv3_act (Activation (None, 19, 19, 728)  0           block7_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv3 (SeparableConv2 (None, 19, 19, 728)  536536      block7_sepconv3_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block7_sepconv3_bn (BatchNormal (None, 19, 19, 728)  2912        block7_sepconv3[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "add_65 (Add)                    (None, 19, 19, 728)  0           block7_sepconv3_bn[0][0]         \n",
+      "                                                                 add_64[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv1_act (Activation (None, 19, 19, 728)  0           add_65[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv1 (SeparableConv2 (None, 19, 19, 728)  536536      block8_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv1_bn (BatchNormal (None, 19, 19, 728)  2912        block8_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv2_act (Activation (None, 19, 19, 728)  0           block8_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv2 (SeparableConv2 (None, 19, 19, 728)  536536      block8_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv2_bn (BatchNormal (None, 19, 19, 728)  2912        block8_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv3_act (Activation (None, 19, 19, 728)  0           block8_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv3 (SeparableConv2 (None, 19, 19, 728)  536536      block8_sepconv3_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block8_sepconv3_bn (BatchNormal (None, 19, 19, 728)  2912        block8_sepconv3[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "add_66 (Add)                    (None, 19, 19, 728)  0           block8_sepconv3_bn[0][0]         \n",
+      "                                                                 add_65[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv1_act (Activation (None, 19, 19, 728)  0           add_66[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv1 (SeparableConv2 (None, 19, 19, 728)  536536      block9_sepconv1_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv1_bn (BatchNormal (None, 19, 19, 728)  2912        block9_sepconv1[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv2_act (Activation (None, 19, 19, 728)  0           block9_sepconv1_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv2 (SeparableConv2 (None, 19, 19, 728)  536536      block9_sepconv2_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv2_bn (BatchNormal (None, 19, 19, 728)  2912        block9_sepconv2[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv3_act (Activation (None, 19, 19, 728)  0           block9_sepconv2_bn[0][0]         \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv3 (SeparableConv2 (None, 19, 19, 728)  536536      block9_sepconv3_act[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block9_sepconv3_bn (BatchNormal (None, 19, 19, 728)  2912        block9_sepconv3[0][0]            \n",
+      "__________________________________________________________________________________________________\n",
+      "add_67 (Add)                    (None, 19, 19, 728)  0           block9_sepconv3_bn[0][0]         \n",
+      "                                                                 add_66[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv1_act (Activatio (None, 19, 19, 728)  0           add_67[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv1 (SeparableConv (None, 19, 19, 728)  536536      block10_sepconv1_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv1_bn (BatchNorma (None, 19, 19, 728)  2912        block10_sepconv1[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv2_act (Activatio (None, 19, 19, 728)  0           block10_sepconv1_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv2 (SeparableConv (None, 19, 19, 728)  536536      block10_sepconv2_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv2_bn (BatchNorma (None, 19, 19, 728)  2912        block10_sepconv2[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv3_act (Activatio (None, 19, 19, 728)  0           block10_sepconv2_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv3 (SeparableConv (None, 19, 19, 728)  536536      block10_sepconv3_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block10_sepconv3_bn (BatchNorma (None, 19, 19, 728)  2912        block10_sepconv3[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "add_68 (Add)                    (None, 19, 19, 728)  0           block10_sepconv3_bn[0][0]        \n",
+      "                                                                 add_67[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv1_act (Activatio (None, 19, 19, 728)  0           add_68[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv1 (SeparableConv (None, 19, 19, 728)  536536      block11_sepconv1_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv1_bn (BatchNorma (None, 19, 19, 728)  2912        block11_sepconv1[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv2_act (Activatio (None, 19, 19, 728)  0           block11_sepconv1_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv2 (SeparableConv (None, 19, 19, 728)  536536      block11_sepconv2_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv2_bn (BatchNorma (None, 19, 19, 728)  2912        block11_sepconv2[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv3_act (Activatio (None, 19, 19, 728)  0           block11_sepconv2_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv3 (SeparableConv (None, 19, 19, 728)  536536      block11_sepconv3_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block11_sepconv3_bn (BatchNorma (None, 19, 19, 728)  2912        block11_sepconv3[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "add_69 (Add)                    (None, 19, 19, 728)  0           block11_sepconv3_bn[0][0]        \n",
+      "                                                                 add_68[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv1_act (Activatio (None, 19, 19, 728)  0           add_69[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv1 (SeparableConv (None, 19, 19, 728)  536536      block12_sepconv1_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv1_bn (BatchNorma (None, 19, 19, 728)  2912        block12_sepconv1[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv2_act (Activatio (None, 19, 19, 728)  0           block12_sepconv1_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv2 (SeparableConv (None, 19, 19, 728)  536536      block12_sepconv2_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv2_bn (BatchNorma (None, 19, 19, 728)  2912        block12_sepconv2[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv3_act (Activatio (None, 19, 19, 728)  0           block12_sepconv2_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv3 (SeparableConv (None, 19, 19, 728)  536536      block12_sepconv3_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block12_sepconv3_bn (BatchNorma (None, 19, 19, 728)  2912        block12_sepconv3[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "add_70 (Add)                    (None, 19, 19, 728)  0           block12_sepconv3_bn[0][0]        \n",
+      "                                                                 add_69[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv1_act (Activatio (None, 19, 19, 728)  0           add_70[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv1 (SeparableConv (None, 19, 19, 728)  536536      block13_sepconv1_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv1_bn (BatchNorma (None, 19, 19, 728)  2912        block13_sepconv1[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv2_act (Activatio (None, 19, 19, 728)  0           block13_sepconv1_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv2 (SeparableConv (None, 19, 19, 1024) 752024      block13_sepconv2_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_sepconv2_bn (BatchNorma (None, 19, 19, 1024) 4096        block13_sepconv2[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "conv2d_23 (Conv2D)              (None, 10, 10, 1024) 745472      add_70[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block13_pool (MaxPooling2D)     (None, 10, 10, 1024) 0           block13_sepconv2_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "batch_normalization_23 (BatchNo (None, 10, 10, 1024) 4096        conv2d_23[0][0]                  \n",
+      "__________________________________________________________________________________________________\n",
+      "add_71 (Add)                    (None, 10, 10, 1024) 0           block13_pool[0][0]               \n",
+      "                                                                 batch_normalization_23[0][0]     \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv1 (SeparableConv (None, 10, 10, 1536) 1582080     add_71[0][0]                     \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv1_bn (BatchNorma (None, 10, 10, 1536) 6144        block14_sepconv1[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv1_act (Activatio (None, 10, 10, 1536) 0           block14_sepconv1_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv2 (SeparableConv (None, 10, 10, 2048) 3159552     block14_sepconv1_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv2_bn (BatchNorma (None, 10, 10, 2048) 8192        block14_sepconv2[0][0]           \n",
+      "__________________________________________________________________________________________________\n",
+      "block14_sepconv2_act (Activatio (None, 10, 10, 2048) 0           block14_sepconv2_bn[0][0]        \n",
+      "__________________________________________________________________________________________________\n",
+      "avg_pool (GlobalAveragePooling2 (None, 2048)         0           block14_sepconv2_act[0][0]       \n",
+      "__________________________________________________________________________________________________\n",
+      "predictions (Dense)             (None, 1000)         2049000     avg_pool[0][0]                   \n",
+      "__________________________________________________________________________________________________\n",
+      "dense_5 (Dense)                 (None, 7)            7007        predictions[0][0]                \n",
+      "==================================================================================================\n",
+      "Total params: 22,917,487\n",
+      "Trainable params: 22,862,959\n",
+      "Non-trainable params: 54,528\n",
+      "__________________________________________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "with mirrored_strategy.scope(): # for training on dual gpus\n",
+    "#    physical_devices = tf.config.list_physical_devices('GPU')\n",
+    "#    tf.config.experimental.set_memory_growth(physical_devices[0], True)\n",
+    "    base_model = tf.keras.applications.xception.Xception(include_top=True, pooling='avg')\n",
+    "    for layer in base_model.layers:\n",
+    "        layer.trainable = True\n",
+    "    output = Dense(7, activation='softmax')(base_model.output)\n",
+    "    model = tf.keras.Model(base_model.input, output)\n",
+    "    model.compile(optimizer=Adam(learning_rate=.001), loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])\n",
+    "# sparse_categorical_crossentropy\n",
+    "#model = add_regularization(model)\n",
+    "model.summary()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 87,
+   "id": "9cd2ba27",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2022-08-01 23:37:26.275294: W tensorflow/core/grappler/optimizers/data/auto_shard.cc:656] In AUTO-mode, and switching to DATA-based sharding, instead of FILE-based sharding as we cannot find appropriate reader dataset op(s) to shard. Error: Did not find a shardable source, walked to a node which is not a dataset: name: \"FlatMapDataset/_2\"\n",
+      "op: \"FlatMapDataset\"\n",
+      "input: \"TensorDataset/_1\"\n",
+      "attr {\n",
+      "  key: \"Targuments\"\n",
+      "  value {\n",
+      "    list {\n",
+      "    }\n",
+      "  }\n",
+      "}\n",
+      "attr {\n",
+      "  key: \"f\"\n",
+      "  value {\n",
+      "    func {\n",
+      "      name: \"__inference_Dataset_flat_map_flat_map_fn_93447\"\n",
+      "    }\n",
+      "  }\n",
+      "}\n",
+      "attr {\n",
+      "  key: \"output_shapes\"\n",
+      "  value {\n",
+      "    list {\n",
+      "      shape {\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "      }\n",
+      "      shape {\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "        dim {\n",
+      "          size: -1\n",
+      "        }\n",
+      "      }\n",
+      "    }\n",
+      "  }\n",
+      "}\n",
+      "attr {\n",
+      "  key: \"output_types\"\n",
+      "  value {\n",
+      "    list {\n",
+      "      type: DT_FLOAT\n",
+      "      type: DT_FLOAT\n",
+      "    }\n",
+      "  }\n",
+      "}\n",
+      ". Consider either turning off auto-sharding or switching the auto_shard_policy to DATA to shard this dataset. You can do this by creating a new `tf.data.Options()` object then setting `options.experimental_distribute.auto_shard_policy = AutoShardPolicy.DATA` before applying the options object to the dataset via `dataset.with_options(options)`.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/6\n",
+      "INFO:tensorflow:batch_all_reduce: 158 all-reduces with algorithm = nccl, num_packs = 1\n",
+      "INFO:tensorflow:batch_all_reduce: 158 all-reduces with algorithm = nccl, num_packs = 1\n",
+      " 17/782 [..............................] - ETA: 6:35 - loss: 1.9460 - accuracy: 0.1428"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "Input \u001b[0;32mIn [87]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtrain_generator\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      2\u001b[0m \u001b[43m          \u001b[49m\u001b[43msteps_per_epoch\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mtrain_generator\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      3\u001b[0m \u001b[43m          \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mvalidation_generator\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[43m          \u001b[49m\u001b[43mvalidation_steps\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mvalidation_generator\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      5\u001b[0m \u001b[43m          \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m6\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m      6\u001b[0m \u001b[43m          \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/training.py:1100\u001b[0m, in \u001b[0;36mModel.fit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1093\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m trace\u001b[38;5;241m.\u001b[39mTrace(\n\u001b[1;32m   1094\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtrain\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m   1095\u001b[0m     epoch_num\u001b[38;5;241m=\u001b[39mepoch,\n\u001b[1;32m   1096\u001b[0m     step_num\u001b[38;5;241m=\u001b[39mstep,\n\u001b[1;32m   1097\u001b[0m     batch_size\u001b[38;5;241m=\u001b[39mbatch_size,\n\u001b[1;32m   1098\u001b[0m     _r\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m):\n\u001b[1;32m   1099\u001b[0m   callbacks\u001b[38;5;241m.\u001b[39mon_train_batch_begin(step)\n\u001b[0;32m-> 1100\u001b[0m   tmp_logs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43miterator\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1101\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m data_handler\u001b[38;5;241m.\u001b[39mshould_sync:\n\u001b[1;32m   1102\u001b[0m     context\u001b[38;5;241m.\u001b[39masync_wait()\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py:828\u001b[0m, in \u001b[0;36mFunction.__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    826\u001b[0m tracing_count \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexperimental_get_tracing_count()\n\u001b[1;32m    827\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m trace\u001b[38;5;241m.\u001b[39mTrace(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_name) \u001b[38;5;28;01mas\u001b[39;00m tm:\n\u001b[0;32m--> 828\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    829\u001b[0m   compiler \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mxla\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_experimental_compile \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnonXla\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    830\u001b[0m   new_tracing_count \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexperimental_get_tracing_count()\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py:855\u001b[0m, in \u001b[0;36mFunction._call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    852\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock\u001b[38;5;241m.\u001b[39mrelease()\n\u001b[1;32m    853\u001b[0m   \u001b[38;5;66;03m# In this case we have created variables on the first call, so we run the\u001b[39;00m\n\u001b[1;32m    854\u001b[0m   \u001b[38;5;66;03m# defunned version which is guaranteed to never create variables.\u001b[39;00m\n\u001b[0;32m--> 855\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_stateless_fn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m  \u001b[38;5;66;03m# pylint: disable=not-callable\u001b[39;00m\n\u001b[1;32m    856\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_stateful_fn \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    857\u001b[0m   \u001b[38;5;66;03m# Release the lock early so that multiple threads can perform the call\u001b[39;00m\n\u001b[1;32m    858\u001b[0m   \u001b[38;5;66;03m# in parallel.\u001b[39;00m\n\u001b[1;32m    859\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock\u001b[38;5;241m.\u001b[39mrelease()\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/function.py:2942\u001b[0m, in \u001b[0;36mFunction.__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2939\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock:\n\u001b[1;32m   2940\u001b[0m   (graph_function,\n\u001b[1;32m   2941\u001b[0m    filtered_flat_args) \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_maybe_define_function(args, kwargs)\n\u001b[0;32m-> 2942\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mgraph_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_flat\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   2943\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfiltered_flat_args\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcaptured_inputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mgraph_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcaptured_inputs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/function.py:1918\u001b[0m, in \u001b[0;36mConcreteFunction._call_flat\u001b[0;34m(self, args, captured_inputs, cancellation_manager)\u001b[0m\n\u001b[1;32m   1914\u001b[0m possible_gradient_type \u001b[38;5;241m=\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPossibleTapeGradientTypes(args)\n\u001b[1;32m   1915\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (possible_gradient_type \u001b[38;5;241m==\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPOSSIBLE_GRADIENT_TYPES_NONE\n\u001b[1;32m   1916\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m executing_eagerly):\n\u001b[1;32m   1917\u001b[0m   \u001b[38;5;66;03m# No tape is watching; skip to running the function.\u001b[39;00m\n\u001b[0;32m-> 1918\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_build_call_outputs(\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_inference_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1919\u001b[0m \u001b[43m      \u001b[49m\u001b[43mctx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcancellation_manager\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcancellation_manager\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[1;32m   1920\u001b[0m forward_backward \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_select_forward_and_backward_functions(\n\u001b[1;32m   1921\u001b[0m     args,\n\u001b[1;32m   1922\u001b[0m     possible_gradient_type,\n\u001b[1;32m   1923\u001b[0m     executing_eagerly)\n\u001b[1;32m   1924\u001b[0m forward_function, args_with_tangents \u001b[38;5;241m=\u001b[39m forward_backward\u001b[38;5;241m.\u001b[39mforward()\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/function.py:555\u001b[0m, in \u001b[0;36m_EagerDefinedFunction.call\u001b[0;34m(self, ctx, args, cancellation_manager)\u001b[0m\n\u001b[1;32m    553\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m _InterpolateFunctionError(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m    554\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m cancellation_manager \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 555\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m \u001b[43mexecute\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexecute\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    556\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mstr\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msignature\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    557\u001b[0m \u001b[43m        \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_num_outputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    558\u001b[0m \u001b[43m        \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    559\u001b[0m \u001b[43m        \u001b[49m\u001b[43mattrs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    560\u001b[0m \u001b[43m        \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mctx\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    561\u001b[0m   \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    562\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m execute\u001b[38;5;241m.\u001b[39mexecute_with_cancellation(\n\u001b[1;32m    563\u001b[0m         \u001b[38;5;28mstr\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msignature\u001b[38;5;241m.\u001b[39mname),\n\u001b[1;32m    564\u001b[0m         num_outputs\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_num_outputs,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    567\u001b[0m         ctx\u001b[38;5;241m=\u001b[39mctx,\n\u001b[1;32m    568\u001b[0m         cancellation_manager\u001b[38;5;241m=\u001b[39mcancellation_manager)\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/eager/execute.py:59\u001b[0m, in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     57\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m     58\u001b[0m   ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[0;32m---> 59\u001b[0m   tensors \u001b[38;5;241m=\u001b[39m \u001b[43mpywrap_tfe\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mTFE_Py_Execute\u001b[49m\u001b[43m(\u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mop_name\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     60\u001b[0m \u001b[43m                                      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     61\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m core\u001b[38;5;241m.\u001b[39m_NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     62\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "model.fit(x=train_generator,\n",
+    "          steps_per_epoch=len(train_generator),\n",
+    "          validation_data=validation_generator,\n",
+    "          validation_steps=len(validation_generator),\n",
+    "          epochs=6,\n",
+    "          verbose=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "63f791af",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model.save(\"Model_1.h5\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

conf_mx_test.ipynb

@@ -0,0 +1,105 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "99d6b339",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2022-08-01 21:12:17.069258: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.datasets import fetch_openml\n",
+    "import matplotlib as mpl\n",
+    "import matplotlib.pyplot as plt\n",
+    "from sklearn.linear_model import SGDClassifier\n",
+    "from sklearn.model_selection import StratifiedKFold, cross_val_predict, train_test_split, StratifiedShuffleSplit,cross_val_score\n",
+    "from sklearn.base import clone, BaseEstimator\n",
+    "from sklearn.metrics import confusion_matrix, f1_score, precision_score, recall_score, precision_recall_curve, roc_curve, roc_auc_score\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.svm import SVC\n",
+    "from sklearn.multiclass import OneVsRestClassifier\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.neighbors import KNeighborsClassifier\n",
+    "\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import tensorflow as tf\n",
+    "\n",
+    "import joblib"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "20c2c97e",
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "('Keyword argument not understood:', 'keepdims')",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "Input \u001b[0;32mIn [7]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0m new_model \u001b[38;5;241m=\u001b[39m \u001b[43mtf\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mkeras\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodels\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mload_model\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mModel_1.h5\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/saving/save.py:206\u001b[0m, in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile, options)\u001b[0m\n\u001b[1;32m    203\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m load_context\u001b[38;5;241m.\u001b[39mload_context(options):\n\u001b[1;32m    204\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m (h5py \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[1;32m    205\u001b[0m       (\u001b[38;5;28misinstance\u001b[39m(filepath, h5py\u001b[38;5;241m.\u001b[39mFile) \u001b[38;5;129;01mor\u001b[39;00m h5py\u001b[38;5;241m.\u001b[39mis_hdf5(filepath))):\n\u001b[0;32m--> 206\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mhdf5_format\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mload_model_from_hdf5\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfilepath\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    207\u001b[0m \u001b[43m                                            \u001b[49m\u001b[38;5;28;43mcompile\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m    209\u001b[0m   filepath \u001b[38;5;241m=\u001b[39m path_to_string(filepath)\n\u001b[1;32m    210\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(filepath, six\u001b[38;5;241m.\u001b[39mstring_types):\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/saving/hdf5_format.py:183\u001b[0m, in \u001b[0;36mload_model_from_hdf5\u001b[0;34m(filepath, custom_objects, compile)\u001b[0m\n\u001b[1;32m    181\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mNo model found in config file.\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m    182\u001b[0m model_config \u001b[38;5;241m=\u001b[39m json_utils\u001b[38;5;241m.\u001b[39mdecode(model_config\u001b[38;5;241m.\u001b[39mdecode(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mutf-8\u001b[39m\u001b[38;5;124m'\u001b[39m))\n\u001b[0;32m--> 183\u001b[0m model \u001b[38;5;241m=\u001b[39m \u001b[43mmodel_config_lib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodel_from_config\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel_config\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    184\u001b[0m \u001b[43m                                           \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcustom_objects\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    186\u001b[0m \u001b[38;5;66;03m# set weights\u001b[39;00m\n\u001b[1;32m    187\u001b[0m load_weights_from_hdf5_group(f[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel_weights\u001b[39m\u001b[38;5;124m'\u001b[39m], model\u001b[38;5;241m.\u001b[39mlayers)\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/saving/model_config.py:64\u001b[0m, in \u001b[0;36mmodel_from_config\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m     60\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m`model_from_config` expects a dictionary, not a list. \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     61\u001b[0m                   \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mMaybe you meant to use \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     62\u001b[0m                   \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m`Sequential.from_config(config)`?\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m     63\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlayers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m deserialize  \u001b[38;5;66;03m# pylint: disable=g-import-not-at-top\u001b[39;00m\n\u001b[0;32m---> 64\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mdeserialize\u001b[49m\u001b[43m(\u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcustom_objects\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/layers/serialization.py:173\u001b[0m, in \u001b[0;36mdeserialize\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m    162\u001b[0m \u001b[38;5;124;03m\"\"\"Instantiates a layer from a config dictionary.\u001b[39;00m\n\u001b[1;32m    163\u001b[0m \n\u001b[1;32m    164\u001b[0m \u001b[38;5;124;03mArguments:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    170\u001b[0m \u001b[38;5;124;03m    Layer instance (may be Model, Sequential, Network, Layer...)\u001b[39;00m\n\u001b[1;32m    171\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    172\u001b[0m populate_deserializable_objects()\n\u001b[0;32m--> 173\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mgeneric_utils\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdeserialize_keras_object\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    174\u001b[0m \u001b[43m    \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    175\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmodule_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mLOCAL\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mALL_OBJECTS\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    176\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcustom_objects\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    177\u001b[0m \u001b[43m    \u001b[49m\u001b[43mprintable_module_name\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mlayer\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/utils/generic_utils.py:354\u001b[0m, in \u001b[0;36mdeserialize_keras_object\u001b[0;34m(identifier, module_objects, custom_objects, printable_module_name)\u001b[0m\n\u001b[1;32m    351\u001b[0m custom_objects \u001b[38;5;241m=\u001b[39m custom_objects \u001b[38;5;129;01mor\u001b[39;00m {}\n\u001b[1;32m    353\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcustom_objects\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;129;01min\u001b[39;00m arg_spec\u001b[38;5;241m.\u001b[39margs:\n\u001b[0;32m--> 354\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_config\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    355\u001b[0m \u001b[43m      \u001b[49m\u001b[43mcls_config\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    356\u001b[0m \u001b[43m      \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mdict\u001b[39;49m\u001b[43m(\u001b[49m\n\u001b[1;32m    357\u001b[0m \u001b[43m          \u001b[49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m_GLOBAL_CUSTOM_OBJECTS\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mitems\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\n\u001b[1;32m    358\u001b[0m \u001b[43m          \u001b[49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mitems\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    359\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m CustomObjectScope(custom_objects):\n\u001b[1;32m    360\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39mfrom_config(cls_config)\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/functional.py:668\u001b[0m, in \u001b[0;36mFunctional.from_config\u001b[0;34m(cls, config, custom_objects)\u001b[0m\n\u001b[1;32m    652\u001b[0m \u001b[38;5;129m@classmethod\u001b[39m\n\u001b[1;32m    653\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mfrom_config\u001b[39m(\u001b[38;5;28mcls\u001b[39m, config, custom_objects\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[1;32m    654\u001b[0m   \u001b[38;5;124;03m\"\"\"Instantiates a Model from its config (output of `get_config()`).\u001b[39;00m\n\u001b[1;32m    655\u001b[0m \n\u001b[1;32m    656\u001b[0m \u001b[38;5;124;03m  Arguments:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    666\u001b[0m \u001b[38;5;124;03m      ValueError: In case of improperly formatted config dict.\u001b[39;00m\n\u001b[1;32m    667\u001b[0m \u001b[38;5;124;03m  \"\"\"\u001b[39;00m\n\u001b[0;32m--> 668\u001b[0m   input_tensors, output_tensors, created_layers \u001b[38;5;241m=\u001b[39m \u001b[43mreconstruct_from_config\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    669\u001b[0m \u001b[43m      \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    670\u001b[0m   model \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mcls\u001b[39m(inputs\u001b[38;5;241m=\u001b[39minput_tensors, outputs\u001b[38;5;241m=\u001b[39moutput_tensors,\n\u001b[1;32m    671\u001b[0m               name\u001b[38;5;241m=\u001b[39mconfig\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mname\u001b[39m\u001b[38;5;124m'\u001b[39m))\n\u001b[1;32m    672\u001b[0m   connect_ancillary_layers(model, created_layers)\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/functional.py:1275\u001b[0m, in \u001b[0;36mreconstruct_from_config\u001b[0;34m(config, custom_objects, created_layers)\u001b[0m\n\u001b[1;32m   1273\u001b[0m \u001b[38;5;66;03m# First, we create all layers and enqueue nodes to be processed\u001b[39;00m\n\u001b[1;32m   1274\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m layer_data \u001b[38;5;129;01min\u001b[39;00m config[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlayers\u001b[39m\u001b[38;5;124m'\u001b[39m]:\n\u001b[0;32m-> 1275\u001b[0m   \u001b[43mprocess_layer\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlayer_data\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1276\u001b[0m \u001b[38;5;66;03m# Then we process nodes in order of layer depth.\u001b[39;00m\n\u001b[1;32m   1277\u001b[0m \u001b[38;5;66;03m# Nodes that cannot yet be processed (if the inbound node\u001b[39;00m\n\u001b[1;32m   1278\u001b[0m \u001b[38;5;66;03m# does not yet exist) are re-enqueued, and the process\u001b[39;00m\n\u001b[1;32m   1279\u001b[0m \u001b[38;5;66;03m# is repeated until all nodes are processed.\u001b[39;00m\n\u001b[1;32m   1280\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m unprocessed_nodes:\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/functional.py:1257\u001b[0m, in \u001b[0;36mreconstruct_from_config.<locals>.process_layer\u001b[0;34m(layer_data)\u001b[0m\n\u001b[1;32m   1253\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1254\u001b[0m   \u001b[38;5;66;03m# Instantiate layer.\u001b[39;00m\n\u001b[1;32m   1255\u001b[0m   \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlayers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m deserialize \u001b[38;5;28;01mas\u001b[39;00m deserialize_layer  \u001b[38;5;66;03m# pylint: disable=g-import-not-at-top\u001b[39;00m\n\u001b[0;32m-> 1257\u001b[0m   layer \u001b[38;5;241m=\u001b[39m \u001b[43mdeserialize_layer\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlayer_data\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcustom_objects\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1258\u001b[0m   created_layers[layer_name] \u001b[38;5;241m=\u001b[39m layer\n\u001b[1;32m   1260\u001b[0m node_count_by_layer[layer] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mint\u001b[39m(_should_skip_first_node(layer))\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/layers/serialization.py:173\u001b[0m, in \u001b[0;36mdeserialize\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m    162\u001b[0m \u001b[38;5;124;03m\"\"\"Instantiates a layer from a config dictionary.\u001b[39;00m\n\u001b[1;32m    163\u001b[0m \n\u001b[1;32m    164\u001b[0m \u001b[38;5;124;03mArguments:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    170\u001b[0m \u001b[38;5;124;03m    Layer instance (may be Model, Sequential, Network, Layer...)\u001b[39;00m\n\u001b[1;32m    171\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    172\u001b[0m populate_deserializable_objects()\n\u001b[0;32m--> 173\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mgeneric_utils\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdeserialize_keras_object\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    174\u001b[0m \u001b[43m    \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    175\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmodule_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mLOCAL\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mALL_OBJECTS\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    176\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcustom_objects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcustom_objects\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    177\u001b[0m \u001b[43m    \u001b[49m\u001b[43mprintable_module_name\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mlayer\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/utils/generic_utils.py:360\u001b[0m, in \u001b[0;36mdeserialize_keras_object\u001b[0;34m(identifier, module_objects, custom_objects, printable_module_name)\u001b[0m\n\u001b[1;32m    354\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39mfrom_config(\n\u001b[1;32m    355\u001b[0m         cls_config,\n\u001b[1;32m    356\u001b[0m         custom_objects\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mdict\u001b[39m(\n\u001b[1;32m    357\u001b[0m             \u001b[38;5;28mlist\u001b[39m(_GLOBAL_CUSTOM_OBJECTS\u001b[38;5;241m.\u001b[39mitems()) \u001b[38;5;241m+\u001b[39m\n\u001b[1;32m    358\u001b[0m             \u001b[38;5;28mlist\u001b[39m(custom_objects\u001b[38;5;241m.\u001b[39mitems())))\n\u001b[1;32m    359\u001b[0m   \u001b[38;5;28;01mwith\u001b[39;00m CustomObjectScope(custom_objects):\n\u001b[0;32m--> 360\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_config\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcls_config\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    361\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    362\u001b[0m   \u001b[38;5;66;03m# Then `cls` may be a function returning a class.\u001b[39;00m\n\u001b[1;32m    363\u001b[0m   \u001b[38;5;66;03m# in this case by convention `config` holds\u001b[39;00m\n\u001b[1;32m    364\u001b[0m   \u001b[38;5;66;03m# the kwargs of the function.\u001b[39;00m\n\u001b[1;32m    365\u001b[0m   custom_objects \u001b[38;5;241m=\u001b[39m custom_objects \u001b[38;5;129;01mor\u001b[39;00m {}\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/base_layer.py:720\u001b[0m, in \u001b[0;36mLayer.from_config\u001b[0;34m(cls, config)\u001b[0m\n\u001b[1;32m    704\u001b[0m \u001b[38;5;129m@classmethod\u001b[39m\n\u001b[1;32m    705\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mfrom_config\u001b[39m(\u001b[38;5;28mcls\u001b[39m, config):\n\u001b[1;32m    706\u001b[0m   \u001b[38;5;124;03m\"\"\"Creates a layer from its config.\u001b[39;00m\n\u001b[1;32m    707\u001b[0m \n\u001b[1;32m    708\u001b[0m \u001b[38;5;124;03m  This method is the reverse of `get_config`,\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    718\u001b[0m \u001b[38;5;124;03m      A layer instance.\u001b[39;00m\n\u001b[1;32m    719\u001b[0m \u001b[38;5;124;03m  \"\"\"\u001b[39;00m\n\u001b[0;32m--> 720\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/layers/pooling.py:862\u001b[0m, in \u001b[0;36mGlobalPooling2D.__init__\u001b[0;34m(self, data_format, **kwargs)\u001b[0m\n\u001b[1;32m    861\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__init__\u001b[39m(\u001b[38;5;28mself\u001b[39m, data_format\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[0;32m--> 862\u001b[0m   \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mGlobalPooling2D\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[38;5;21;43m__init__\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    863\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdata_format \u001b[38;5;241m=\u001b[39m conv_utils\u001b[38;5;241m.\u001b[39mnormalize_data_format(data_format)\n\u001b[1;32m    864\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minput_spec \u001b[38;5;241m=\u001b[39m InputSpec(ndim\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m4\u001b[39m)\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/training/tracking/base.py:517\u001b[0m, in \u001b[0;36mno_automatic_dependency_tracking.<locals>._method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    515\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_self_setattr_tracking \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m  \u001b[38;5;66;03m# pylint: disable=protected-access\u001b[39;00m\n\u001b[1;32m    516\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 517\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[43mmethod\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    518\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m    519\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_self_setattr_tracking \u001b[38;5;241m=\u001b[39m previous_value  \u001b[38;5;66;03m# pylint: disable=protected-access\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/engine/base_layer.py:340\u001b[0m, in \u001b[0;36mLayer.__init__\u001b[0;34m(self, trainable, name, dtype, dynamic, **kwargs)\u001b[0m\n\u001b[1;32m    329\u001b[0m allowed_kwargs \u001b[38;5;241m=\u001b[39m {\n\u001b[1;32m    330\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124minput_dim\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m    331\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124minput_shape\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    337\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mimplementation\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m    338\u001b[0m }\n\u001b[1;32m    339\u001b[0m \u001b[38;5;66;03m# Validate optional keyword arguments.\u001b[39;00m\n\u001b[0;32m--> 340\u001b[0m \u001b[43mgeneric_utils\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mvalidate_kwargs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mallowed_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    342\u001b[0m \u001b[38;5;66;03m# Mutable properties\u001b[39;00m\n\u001b[1;32m    343\u001b[0m \u001b[38;5;66;03m# Indicates whether the layer's weights are updated during training\u001b[39;00m\n\u001b[1;32m    344\u001b[0m \u001b[38;5;66;03m# and whether the layer's updates are run during training.\u001b[39;00m\n\u001b[1;32m    345\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_trainable \u001b[38;5;241m=\u001b[39m trainable\n",
+      "File \u001b[0;32m~/miniconda3/envs/tensorflow-cuda/lib/python3.9/site-packages/tensorflow/python/keras/utils/generic_utils.py:808\u001b[0m, in \u001b[0;36mvalidate_kwargs\u001b[0;34m(kwargs, allowed_kwargs, error_message)\u001b[0m\n\u001b[1;32m    806\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m kwarg \u001b[38;5;129;01min\u001b[39;00m kwargs:\n\u001b[1;32m    807\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m kwarg \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m allowed_kwargs:\n\u001b[0;32m--> 808\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(error_message, kwarg)\n",
+      "\u001b[0;31mTypeError\u001b[0m: ('Keyword argument not understood:', 'keepdims')"
+     ]
+    }
+   ],
+   "source": [
+    "new_model = tf.keras.models.load_model('Model_1.h5')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "664cf629",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

curate.py

@@ -11,16 +11,16 @@ training = curate.to_training(raw_data) # creates raw_df
 class_training = curate.class_training(training) # creates initial class_training df
 nvl_training = curate.nvl_training(training) # creates initial nvl_training
 dropd = curate.drop_nvl_cols(nvl_training)  # label mask
+dropd
 
-expanded_dfs = curate.expand_nvlclass(class_training, dropd) # pulls values out of lists for both dfs
+# pulls values out of lists for both dfs and creates temp_pics_source_list.txt
+expanded_dfs = curate.expand_nvlclass(class_training, dropd)
 
 expanded_class = expanded_dfs[0] # TODO still having problems with Unnamed: 0 col
 expanded_dropd = expanded_dfs[1] # TODO incorrect df. Look at nvl_training func. Specifically "reindex" usage
 
 download = input('download images?: ')
 if ('y' or 'Y') in download:
-    with open('temp_pics_source_list.txt') as f:
-        test_list = json.load(f)
-    curate.dl_pictures(test_list)
+    curate.dl_pictures()
 else:
     pass

ebay_api.py

@@ -16,12 +16,151 @@ import pandas as pd
 import config as cfg
 import shutil
 import re
+import urllib, base64
 
 from ebaysdk.exception import ConnectionError
 from ebaysdk.trading import Connection as Trading
 from ebaysdk.finding import Connection as Finding
 from ebaysdk.shopping import Connection as Shopping
 
+# renew oauth token for shopping api
+def getAuthToken():
+     AppSettings = {
+          'client_id': cfg.oauth["client_id"],
+          'client_secret':cfg.oauth["client_secret"],
+          'ruName':cfg.oauth["RuName"]
+          }
+
+     authHeaderData = AppSettings['client_id'] + ':' + AppSettings['client_secret']
+     encodedAuthHeader = base64.b64encode(str.encode(authHeaderData))
+     encodedAuthHeader = str(encodedAuthHeader)[2:len(str(encodedAuthHeader))-1]
+
+     headers = {
+          "Content-Type" : "application/x-www-form-urlencoded", # what is this?
+          "Authorization" : "Basic " + str(encodedAuthHeader)
+          }
+
+     body= {
+          "grant_type" : "client_credentials",
+          "redirect_uri" : AppSettings['ruName'],
+          "scope" : "https://api.ebay.com/oauth/api_scope"
+      }
+
+     data = urllib.parse.urlencode(body)
+
+     tokenURL = "https://api.ebay.com/identity/v1/oauth2/token"
+
+     response = requests.post(tokenURL, headers=headers, data=data).json()
+#     error = response['error_description'] #if errors
+     access_token = response['access_token']
+
+     with open('temp_oauth_token.txt', 'w') as f:
+         json.dump(access_token, f)
+
+     return access_token
+
+class FindingApi:
+    '''
+    Methods for accessing eBay's FindingApi services
+    '''
+
+    def __init__(self, service):
+        self.service = [
+            'findItemsAdvanced', 'findCompletedItems',
+            'findItemsByKeywords', 'findItemsIneBayStores', 'findItemsByCategory',
+            'findItemsByProduct'
+            ][service] # Currently using only index 4, i.e., service = 4
+
+    def get_data(self, category_id):
+
+        '''
+        Gets raw JSON data fom FindingApi service call. Currently being used to
+        get itemIDs from categories;
+        '''
+#        startTime = dateutil.parser.isoparse( startTime )
+#        now = datetime.datetime.now(tz=pytz.UTC)
+#        days_on_site = (now - startTime).days # as int
+
+        ids = []
+        params = {
+            "OPERATION-NAME":self.service,
+            "SECURITY-APPNAME":cfg.sec['SECURITY-APPNAME'],
+            "SERVICE-VERSION":"1.13.0",
+            "RESPONSE-DATA-FORMAT":"JSON",
+            "categoryId":category_id,
+            "paginationInput.entriesPerPage":"100",
+            "paginationInput.PageNumber":"1",
+            "itemFilter(0).name":"Condition",
+            "itemFilter(0).value":"Used",
+            "itemFilter.name":"HideDuplicateItems",
+            "itemFilter.value":"true",
+            "sortOrder":"StartTimeNewest",
+            }
+#            "itemFilter(1).name":"TopRatedSellerOnly", # TODO fix here
+#            "itemFilter(1).value":"true"
+
+        try:
+            response = requests.get("https://svcs.ebay.com/services/search/FindingService/v1",
+                params=params, timeout=24)
+            response.raise_for_status()
+
+        except requests.exceptions.RequestException: # appears this works need to be able to continue where you left off or use better timeout?
+            print('connection error')
+            return ids
+        try:
+            data = response.json()
+            for item in data['findItemsByCategoryResponse'][0]['searchResult'][0]['item']:
+                ids.append(item['itemId'][0])
+
+            ids = list(set(ids))
+
+        except (AttributeError, KeyError):
+            print('AttributeError or KeyError. Exiting')
+            print(response.json())
+            return ids
+
+        return ids
+
+# TODO add some other options to finding call api such as for possibly filtering for used items only. This might give you a better dataset for training. Or maybe a mixture of new and used. Maybe 
+# try and come up with a way to mathematically determine your odds of maximizing the number of pictures in your training set while reducing the number of useless images. Say for example, if you took a
+# random set of 3 of 8 pictures total from each listing you might have a better chance of getting 3 good pictures in addition to increasing your training set. Or maybe you would have better luck with limiting
+# it to the first 5 pictures instead of random. 
+
+# You may even have more consistency with used shoes since they are "one-off" items without confusing multiple variations and colors. What else you can do is run small training sets on both new and used
+# to see which one is more accurate or if a combo of both is more accurate. 
+
+    def get_ids_from_cats(self):
+        '''
+        Creates a 20-itemId list to use for the ShoppingApi
+        call
+        '''
+
+        ids = []
+
+        # load category id list
+        with open('cat_list.txt') as jf:
+            cat_list = json.load(jf)
+
+        # load list of master ids
+        with open('master_ids.txt') as f:
+            master_ids = json.load(f)
+
+        # fetch ids with calls to Finding Api given cats as param
+        with concurrent.futures.ThreadPoolExecutor() as executor:
+            for future in executor.map(self.get_data, cat_list):
+                ids.extend(future)
+
+        # append master ids list with temporary ids from single function call and save
+        master_ids.extend(ids)
+        master_ids = list(set(master_ids))
+        with open('master_ids.txt', 'w') as f:
+            json.dump(master_ids, f)
+
+ # 20-ItemID list created to maximize dataset/decrease calls provided call constraints
+        twenty_id_list = [','.join(ids[n:n+20]) for n in list(range(0,
+            len(ids), 20))]
+
+        return twenty_id_list, ids
 
 class ShoppingApi:
     '''
@@ -29,6 +168,14 @@ class ShoppingApi:
     pandas dataframes
     '''
 
+#    def __init__(self):
+#        
+#        # renew oauth token
+#        oauth_response = getAuthToken()
+#        access_token = oauth_response[0]
+#        
+#        self.access_token = access_token
+
     def update_cats(self):
         '''
         Updates cat_list.txt
@@ -37,19 +184,19 @@ class ShoppingApi:
         parent_cats = ['3034', '93427'] # Women's and Men's shoe departments
         cat_list = []
 
+        with open('temp_oauth_token.txt') as f:
+            access_token = json.load(f)
         for department in parent_cats:
 
-            params = {
-                "callname":"GetCategoryInfo",
-                "X-EBAY-API-IAF-TOKEN":cfg.sec['X-EBAY-API-IAF-TOKEN'],
+            headers = {
+                "X-EBAY-API-IAF-TOKEN":access_token,
                 "version":"671",
-                "responseencoding":"JSON",
-                "CategoryID":department,
-                "IncludeSelector":"ChildCategories",
                 }
 
+            url = "https://open.api.ebay.com/shopping?&callname=GetCategoryInfo&responseencoding=JSON&IncludeSelector=ChildCategories&CategoryID="+department
+
             try:
-                response = requests.get("https://open.api.ebay.com/shopping?", params=params, timeout=4)
+                response = requests.get(url, headers=headers, timeout=4)
                 response.raise_for_status()
 
             except requests.exceptions.RequestException:
@@ -59,49 +206,52 @@ class ShoppingApi:
             response = response['CategoryArray']['Category'][1:] # excludes index 0 as this is parent node, i.e., women's or men's dept.
 
             temp_cat_list = [cat['CategoryID'] for cat in response]
+
+            if department == '3034':
+                women_cats = temp_cat_list
+            elif department == '93427':
+                men_cats = temp_cat_list
+
             cat_list.extend(temp_cat_list)
 
-            with open('cat_list.txt', 'w') as f:
-                json.dump(cat_list, f)
-
-            # leaf_list = [node['LeafCategory'] for node in response]
+        with open('cat_list.txt', 'w') as f:
+            json.dump(cat_list, f)
+        with open('women_cat_list.txt', 'w') as f:
+            json.dump(women_cats, f)
+        with open('men_cat_list.txt', 'w') as f:
+            json.dump(men_cats, f)
 
     def get_item_from_findItemsByCategory(self, twenty_id):
 
         '''
-        Gets raw JSON data from multiple live listings given multiple itemIds
+       Gets raw JSON data from multiple live listings given multiple itemIds
         '''
-
-        with open('item_id_results.txt') as f:
-            item_id_results = json.load(f)
+        with open('temp_oauth_token.txt') as f:
+            access_token = json.load(f)
 
         headers = {
-            "X-EBAY-API-IAF-TOKEN":cfg.sec['X-EBAY-API-IAF-TOKEN'], # TODO implement auto oauth token renewal
+            "X-EBAY-API-IAF-TOKEN":access_token,
             "version":"671",
             }
 
         url = "https://open.api.ebay.com/shopping?&callname=GetMultipleItems&responseencoding=JSON&IncludeSelector=ItemSpecifics&ItemID="+twenty_id
 
         try:
-            # random sleep here between 0 and 10 secs?
         
-            sleep(randint(1,10)) # may not be necessary
             response = requests.get(url, headers=headers,timeout=24)
             response.raise_for_status()
             response = response.json()
-            response = response['Item']
-            print('index number {}'.format(item_id_results.index(twenty_id)))
-            print(response)
+            item = response['Item']
 
 
         except (requests.exceptions.RequestException, KeyError):
             print('connection error. IP limit possibly exceeded')
-            print('index number {}'.format(item_id_results.index(twenty_id)))
-            return # returns NoneType. Handle at conky()
+            print(response)
+            return # this returns NoneType. Handled at conky()
 
-        return response
+        return item
 
-    def conky(self):
+    def conky(self, twenty_ids_list):
         '''
         Runs get_item_from_findItemsByCategory in multiple threads to get relevant
         data for creating training sets
@@ -112,23 +262,17 @@ class ShoppingApi:
         except (FileNotFoundError, ValueError):
             data = []
 
-        try:
-            with open('item_id_results.txt') as f:
-                item_id_results = json.load(f)
-
-        except (FileNotFoundError, ValueError):
-            item_id_results = scrape_ids.main()
-
         with concurrent.futures.ThreadPoolExecutor() as executor:
-            for future in executor.map(self.get_item_from_findItemsByCategory, item_id_results):
+            for future in executor.map(self.get_item_from_findItemsByCategory, twenty_ids_list):
                 if future is not None:
                     for item in future:
                         data.append(item) # The end result should be a list of dicts where each dict in the list is a listing 
                 else:
-                    print('reached call limit')
+                    print('response is None')
                     break
         with open('raw_data.txt', 'w') as f:
             json.dump(data, f)
+        return data
 
 # NOTE:
 
@@ -262,74 +406,106 @@ class CurateData:
         '''
         expand = input("expand image list or use primary listing image? (y or n): ")
         if ('y' or 'Y') in expand:
-            expanded_class = class_training.explode('PictureURL').reset_index(drop=True)
-            expanded_class = expanded_class.dropna(subset=['PictureURL'])
-            expanded_class = expanded_class.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
+            count = input('how many images? All [A] or the first <n> images?')
+            if 'A' in count:
+                expanded_class = class_training.explode('PictureURL').reset_index(drop=True)
+                expanded_class = expanded_class.dropna(subset=['PictureURL'])
+                expanded_class = expanded_class.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
 
-            expanded_dropd = dropd.explode('PictureURL').reset_index(drop=True)
-            expanded_dropd = expanded_dropd.dropna(subset=['PictureURL'])
-            expanded_dropd = expanded_dropd.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
+                expanded_dropd = dropd.explode('PictureURL').reset_index(drop=True)
+                expanded_dropd = expanded_dropd.dropna(subset=['PictureURL'])
+                expanded_dropd = expanded_dropd.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
 
-            expanded_dropd = self.extract_df(expanded_dropd) # convert lists to values
+                expanded_dropd = self.extract_df(expanded_dropd) # convert lists to values
 
-            temp_pics_source_list = list(set(expanded_class.PictureURL.to_list())) # TODO because var is del after dl_pictures you may be
-            # getting duplicate pictures. ie, expanded_class.PictureURL is a master series and will write temp_pics_source_list as such
-            # giving you many repeated pictureURLs (they will not get downloaded due to check @ dl_pic but checking will cont to grow in 
-            # computate power reqs. So, figure out a way to make a true temp list based on the current call executed
+                temp_pics_source_list = list(set(expanded_class.PictureURL.to_list())) 
+            else:
+                count = int(count)
+                class_training['PictureURL'] = class_training['PictureURL'].apply(lambda x: x[0:count] if len(x)>0 else np.nan)
+                expanded_class = class_training.explode('PictureURL').reset_index(drop=True)
+                expanded_class = expanded_class.dropna(subset=['PictureURL'])
+                expanded_class = expanded_class.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
+
+                dropd = dropd.dropna(subset=['PictureURL'])
+                dropd['PictureURL'] = dropd['PictureURL'].apply(lambda x: x[0:count] if len(x)>0 else np.nan)
+                expanded_dropd = dropd.explode('PictureURL').reset_index(drop=True)
+                expanded_dropd = expanded_dropd.dropna(subset=['PictureURL'])
+
+                expanded_dropd = self.extract_df(expanded_dropd) # convert lists to values
+
+                # retrieves picture URLs from master raw_data.txt and rewrites temp_pics_source_list.txt
+                temp_pics_source_list = list(set(expanded_class.PictureURL.to_list())) # TODO
 
         else:
-            class_training['PictureURL'] = class_training['PictureURL'].apply(lambda x: x[0])
-            expanded_class = class_training
-            dropd['PictureURL'] = dropd['PictureURL'].apply(lambda x: x[0])
+            class_training['PictureURL'] = class_training['PictureURL'].apply(lambda x: x[0] if len(x)>0 else np.nan)
+            expanded_class = class_training.dropna()
+            dropd = dropd.dropna(subset=['PictureURL'])
+            dropd['PictureURL'] = dropd['PictureURL'].apply(lambda x: x[0] if len(x)>0 else np.nan)
+            dropd = dropd.dropna(subset=['PictureURL'])
             expanded_dropd = dropd
 
             expanded_dropd = self.extract_df(expanded_dropd) # convert lists to values
+
+            # retrieves picture URLs from master raw_data.txt and rewrites temp_pics_source_list.txt
             temp_pics_source_list = list(set(expanded_class.PictureURL.to_list()))
 
         try:
             with open('temp_pics_source_list.txt') as f:
                 tpsl = json.load(f)
                 tpsl.extend(temp_pics_source_list)
+
+                # ensures no duplicate source URLs exist
                 temp_pics_source_list = list(set(tpsl))
                 with open('temp_pics_source_list.txt', 'w') as f:
                     json.dump(temp_pics_source_list, f)
+
+        # creates file if script is ran for 1st time and file not present
         except (ValueError, FileNotFoundError):
             with open('temp_pics_source_list.txt', 'w') as f:
                 json.dump(temp_pics_source_list, f)
 
         # Append to master training dataframes, drop potential dupes and save
-
         expanded_class.to_csv('expanded_class.csv')
-        # expanded_class = pd.read_csv('expanded_class.csv', index_col=0)
-        # expanded_class.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
-        # expanded_class.to_csv('expanded_class.csv', mode='a', encoding='utf-8') # TODO see line 235 about views and copies
-
         expanded_dropd.to_csv('expanded_dropd.csv')
-        # expanded_dropd = pd.read_csv('expanded_dropd.csv', index_col=0)
-        # expanded_dropd.drop_duplicates(subset=['PictureURL']).reset_index(drop=True)
-        # expanded_dropd.to_csv('expanded_dropd.csv', mode='a', encoding='utf-8')
 
         return expanded_class, expanded_dropd
 
-    def dl_pictures(self, *args):
-        '''
-        Downloads pictures from api to local storage using temp_pics_source_list
-        and creates custom {source:target} dictionary as dict_pics
-        '''
+    def dl_pic(self,dict_pics, pic):
 
-        # TODO add option to include only first image of each listing as
-        # others may be crappy for training. Also consider adding option to
-        # reduce the size of each pic downloaded
+        try:
+
+            # check if image exists in current working directory. avoids dupes
+            if os.path.exists(dict_pics[pic]):
+                pass
+
+            else:
+
+                try:
+
+                    r = requests.get(pic, stream=True)
+                    r.raw.decode_content = True
+                    with open(dict_pics[pic], 'wb') as  f:
+                        shutil.copyfileobj(r.raw, f)
+
+                except ConnectionError:
+                    return
+
+        except KeyError:
+            pass
+
+    def dict_pics(self):
 
         try:
             with open('target_dirs.txt', 'r+') as f: # TODO you can add option to change directory here, too. Look up how to have optional arguments
                 target_dir = json.load(f)
+
         except (ValueError, FileNotFoundError):
             target_dir = input('No target dirctory found. Create One? [y] or [n]:')
             if target_dir == ('y' or 'Y'):
                 target_dir = input('Please provide full URL to destination folder:') # TODO need to catch human syntax errors here
                 with open('target_dirs.txt','w') as f:
                     json.dump(target_dir, f)
+                
             else:
                 os.mkdir(os.getcwd()+os.sep+'training_images')
                 target_dir = os.getcwd()+os.sep+'training_images'
@@ -337,58 +513,59 @@ class CurateData:
                     json.dump(target_dir, f)
                     print('Creating default folder in current directory @ ' + target_dir)
 
+        # open url list in working directory
+        with open('temp_pics_source_list.txt') as f:
+            
+            try:
+                temp_pics_source_list = json.load(f)
+
+            except (ValueError, FileNotFoundError):
+                print('url list not found. aborting')
+                return
+
+        dict_pics = {}
+
+        # make custom dict, {source:target}, and name images from unique URL patt
+        for k in temp_pics_source_list:
+            try:
+                patt_1 = re.search(r'[^/]+(?=/\$_|.(\.jpg|\.jpeg|\.png))', k, re.IGNORECASE)
+                patt_2 = re.search(r'(\.jpg|\.jpeg|\.png)', k, re.IGNORECASE) 
+                if patt_1 and patt_2 is not None:
+                    tag = patt_1.group() + patt_2.group().lower()
+                    file_name = target_dir + os.sep + tag
+                    dict_pics.update({k:file_name})
+            except TypeError:
+                pass
+
+        with open('dict_pics.txt', 'w') as f:
+            json.dump(dict_pics, f)
+
+        return dict_pics # TODO still need to find sol to outliers (aka, naming scheme for unusual source URLs)
+
+    def dl_pictures(self, *dict_pics):
+        '''
+        Downloads pictures from api to local storage using temp_pics_source_list
+        and dict_pics
+        '''
+
+        if not dict_pics:
+            dict_pics = self.dict_pics()
+
         with open('temp_pics_source_list.txt') as f:
             try:
-                if args:
-                    temp_pics_source_list = args[0]
-                else:
-                    temp_pics_source_list = json.load(f)
+                temp_pics_source_list = json.load(f)
             except (ValueError, FileNotFoundError):
-                if args:
-                    temp_pics_sources_list = args[0]
-                else:
-                    print('url list not found. download aborted')
-                    return
-
-        temp_dict_pics = {k:target_dir + os.sep + re.search(r'[^/]+(?=/\$_|.jpg)', k, re.IGNORECASE).group() + '.jpg' for k in temp_pics_source_list}
-
-        try:
-            with open('dict_pics.txt') as f:
-                dict_pics = json.load(f)
-                dict_pics.update(temp_dict_pics) # TODO This still creates duplicates
-            with open('dict_pics.txt', 'w') as f:
-                json.dump(dict_pics, f)
-
-        except (ValueError, FileNotFoundError):
-            with open('dict_pics.txt', 'w') as f:
-                json.dump(temp_dict_pics, f)
-                dict_pics = temp_dict_pics
-
-        def dl_pic(dict_pics, pic):
-
-            if os.path.exists(dict_pics[pic]): # or call temp_dict_pics[pic] can work
-                pass # TODO This is not catching duplicates for some reason....possibly not? Upon inspection, files aren't duplicates...but why?
-            #TODO it would mean that temp_pics_source_list is changing for some reason?
-
-            else:
-                try:
-                    r = requests.get(pic, stream=True)
-                    r.raw.decode_content = True
-                    with open(temp_dict_pics[pic], 'wb') as  f: # Or call dict_pics[pic] can work
-                        shutil.copyfileobj(r.raw, f)
-                except ConnectionError:
-                    return
+                print('url list not found. download aborted')
+                return
 
         bargs = [(dict_pics, pic) for pic in temp_pics_source_list]
         with concurrent.futures.ThreadPoolExecutor() as executor:
-            for future in executor.map(lambda p: dl_pic(*p), bargs):
+            for future in executor.map(lambda p: self.dl_pic(*p), bargs):
                 if future is not None:
                     future
                 else:
                     print('connection error')
 
-        os.remove('temp_pics_source_list.txt') # Deletes file after downloads complete successfully
-
 class PreProcessing:
     '''
     Includes methods for pre-processing training set input and labels in the
@@ -400,13 +577,18 @@ class PreProcessing:
     splits, etc.
     '''
 
-    def stt_training(self, dict_pics, expanded_class, expanded_dropd):
+    def dict_pics(self):
         '''
         Source to target training. Replaces source image URL with target URL
         determined by values in dict_pics variable.
         '''
         
-        pass
+        target_dir = os.getcwd()
+        with open('temp_pics_source_list.txt') as f:
+            temp_pics_source_list = json.load(f)
+        dict_pics = {k:target_dir + os.sep + re.search(r'[^/]+(?=/\$_|.jpg)', k, re.IGNORECASE).group() + '.jpg' for k in temp_pics_source_list}
+        print("{source:target} dictionary created @ " + os.getcwd() + os.sep + 'training_images')
+        return dict_pics
 
         # TODO pipeline gameplan: 5 files: dict_pics.txt,raw_json.txt, raw_json.csv, expanded_class.csv, expanded_dropd.csv
         # cont... open raw_json.txt and append, same with csv --> process new data --> pull out image source+dest and expand new dfs for the additional pictures

image_faults.py

@@ -0,0 +1,28 @@
+import os
+import PIL
+from pathlib import Path
+from PIL import UnidentifiedImageError, Image
+
+'''
+Since PIL is used in keras to open images, you need to identify and remove
+faulty images to avoid hiccups in training. When these are removed from their
+parent folders, their corresponding row in the dataframe should also be removed.
+But because the dataframe is constructed as such:
+    
+'''
+def faulty_images():
+    path = Path("training_images").rglob("*.jpg")
+    for img_p in path:
+        try:
+            img = PIL.Image.open(img_p)
+        except PIL.UnidentifiedImageError:
+                os.remove(img_p)
+#                print(img_p + "Removed")
+#             remove from folder, dataset(is constructed from the csv files
+#             ), dict_pics, temp_pics_source_list,
+#             expanded_dropd, expanded_class. But, remember that if you run curate.py
+#             again the same faulty images will be recreated since it's still in
+#             the raw_data.txt file
+
+if __name__=="__main__":
+    faulty_images()

rand_revise.py

@@ -0,0 +1,103 @@
+import ebaysdk
+import json
+import requests
+import random
+from ebaysdk.trading import Connection as Trading
+from ebaysdk.finding import Connection as Finding
+from ebaysdk.shopping import Connection as Shopping
+import concurrent.futures
+import config as cfg
+import store_ids
+import ebay_api
+
+tapi = Trading(config_file='ebay.yaml')
+
+
+def revised_price(id, original_prices):
+    percent = (random.randint(95, 105))/100
+    rev_price = original_prices[id]*percent
+    rev_price = str(round(rev_price, 2))
+    return rev_price
+
+def revise_item(id, rev_price):
+    response = tapi.execute(
+        'ReviseItem', {
+            'item': {
+            'ItemID': id,
+            'StartPrice':rev_price
+            }
+        
+        }
+    )
+
+def revise_items():
+    with open('original_prices.txt') as f:
+        original_prices = json.load(f)
+
+    for id in original_prices:
+        rev_price = revised_price(id, original_prices)
+        revise_item(id, rev_price)
+
+def get_prices(twenty_id):
+
+    '''
+   Gets raw JSON data from multiple live listings given multiple itemIds
+    '''
+
+    with open('temp_oauth_token.txt') as f:
+        access_token = json.load(f)
+
+    headers = {
+        "X-EBAY-API-IAF-TOKEN":access_token,
+        "version":"671",
+        }
+
+    url = "https://open.api.ebay.com/shopping?&callname=GetMultipleItems&responseencoding=JSON&ItemID="+twenty_id
+
+    try:
+    
+        response = requests.get(url, headers=headers,timeout=24)
+        response.raise_for_status()
+        response = response.json()
+        item = response['Item']
+
+
+    except (requests.exceptions.RequestException, KeyError):
+        print('connection error. IP limit possibly exceeded')
+        print(response)
+        return # this returns NoneType. Handled at get_prices_thread
+
+    id_price_dict = {item['ItemID']:item['ConvertedCurrentPrice']['Value'] for item in item}
+    return id_price_dict
+
+def get_prices_thread(twenty_ids_list):
+    '''
+    Runs get_prices in multiple threads
+    '''
+
+    id_price_dict = {}
+
+    with concurrent.futures.ThreadPoolExecutor() as executor:
+        for future in executor.map(get_prices, twenty_ids_list):
+            if future is not None:
+                id_price_dict.update(future)
+            else:
+                print('response is None')
+                break
+    return id_price_dict
+
+def main():
+
+    with open('ebay_ids.txt') as f:
+        ids = json.load(f)
+
+    twenty_id_list = [','.join(ids[n:n+20]) for n in list(range(0,
+        len(ids), 20))]
+
+    ids = store_ids.main() # gets your store ids for all listings
+    ebay_api.getAuthToken() # updates your Oauth Token
+    id_price_dict = get_prices_thread(twenty_id_list)
+    return id_price_dict
+
+if __name__=="__main__":
+    main()

scrape_ids.py

@@ -36,15 +36,15 @@ def get_isurl(category_id): # "get itemSearchURL"
         return url
     try:
         data = response.json()
+        print(data)
         # NOTE approx 220 pages of listings per cat @ 35 items per page
         item_cond = "&rt=nc&LH_ItemCondition=3000&mag=1" # preowned
+        item_cond_new = '&LH_ItemCondition=3'
         urls = []
-        url = data['findItemsByCategoryResponse'][0]['itemSearchURL'][0] 
-        url = url+item_cond
-        j = list(range(1,221))
-        for i in j:
-            pg = "&_pgn={}".format(str(i))
-            url = url.replace('&_pgn=1', pg)
+        base_url = data['findItemsByCategoryResponse'][0]['itemSearchURL'][0] 
+        for pg in list(range(1,34)): # No results after around page 32
+            url = base_url+"&_pgn="+str(pg)+item_cond
+            print(url)
             urls.append(url)
             
     except (AttributeError, KeyError):
@@ -70,17 +70,22 @@ def get_ids(url):
     '''
     html = requests.get(url).text
     soup = b(html, "html.parser")
+    print(soup)
     ids = list(soup.find_all(href=re.compile(r"[\d]+(?=\?hash)")))
     ids = [id['href'] for id in ids]
     ids = [re.findall(r"[\d]+(?=\?)", id)[0] for id in ids]
-    ids = list(set(ids)) # necessary; two links are returned with pattern match
+    print(ids)
 
     return ids
 
 def threaded_get_ids(urls):
-    
+    '''
+    Runs get_ids() w/in ThreadPoolExecutor() for multi threaded requests.
+    Constructs and saves unique ids and 20_itemIDs for use with ebay_api
+    methods
+   ''' 
     try:
-        with open('item_id_results.txt') as f:
+        with open('ids.txt') as f:
             ids = json.load(f)
     except FileNotFoundError:
         ids = []
@@ -89,14 +94,32 @@ def threaded_get_ids(urls):
         for future in executor.map(get_ids, urls):
             ids.extend(future)
 
+    ids = list(set(ids)) # necessary; two links are returned with pattern match
     item_id_results = [','.join(ids[n:n+20]) for n in list(range(0,
     len(ids), 20))] # 20-ItemID list created to maximize dataset/decrease calls given call constraints
 
+    with open('ids.txt', 'w') as f:
+        json.dump(ids,f)
+
     with open('item_id_results.txt', 'w') as f:
         json.dump(item_id_results, f)
 
     return item_id_results
 
+def  id_count():
+    '''
+    Counts Unique IDs of item_id_results for testing
+    '''
+    with open('item_id_results.txt') as f:
+        item_id_results = json.load(f)
+    
+    ids = ','.join(item_id_results)
+    ids = ids.split(',')
+    uniq = len(list(set(ids)))
+    print('{} Unique IDs'.format(uniq))
+
+    return ids
+
 def main():
     urls = threaded_urls()
     item_id_results = threaded_get_ids(urls)

shopping.py

@@ -1,7 +0,0 @@
-'''
-Initial download and write of raw data from ebay
-'''
-import ebay_api
-
-shopping = ebay_api.ShoppingApi()
-data = shopping.conky()

store_ids.py

@@ -0,0 +1,70 @@
+import os
+import requests
+import json
+import ebaysdk
+from ebaysdk.trading import Connection as Trading
+from ebaysdk.finding import Connection as Finding
+import time
+import concurrent.futures
+# (categoryId = women's shoes = 3034)
+# Initialize loop to get number of pages needed in for loop
+start = time.time()
+fapi = Finding(config_file = "ebay.yaml")
+tapi = Trading(config_file = 'ebay.yaml')
+
+fresponse = fapi.execute(
+        'findItemsAdvanced',
+        {
+            'itemFilter':{
+                'name':'Seller',
+                'value':'chesshoebuddy'
+                },
+            'paginationInput':{
+                'entriesPerPage':'100',
+                'pageNumber':'1'
+                }
+            }
+        ).dict()
+
+page_results = int(fresponse['paginationOutput']['totalPages'])
+
+pages = []
+for i in range(0, page_results):
+    i += 1
+    pages.append(i)
+
+''' Begin definitions for getting ItemIds and SKU: '''
+
+def id_up(n):
+    ids = []
+    fresponse = fapi.execute(
+            'findItemsAdvanced',
+            {
+                'itemFilter':{
+                    'name':'Seller',
+                    'value':'chesshoebuddy'
+                    },
+                'paginationInput':{
+                    'entriesPerPage':'100',
+                    'pageNumber':str(n)
+                    }
+                }
+            ).dict()
+    for item in (fresponse['searchResult']['item']):
+        itemID = item['itemId']
+        #response = tapi.execute('GetItem',{'ItemID':itemID}).dict()
+        ids.append(itemID)
+    return ids
+
+def main():
+    ids = []
+    skus = []
+    with concurrent.futures.ThreadPoolExecutor() as executor:
+        for future in executor.map(id_up, pages):
+            ids.extend(future)
+
+    with open('ebay_ids.txt', 'w') as outfile:
+        json.dump(ids, outfile)
+
+if __name__ == '__main__':
+    main()

testing.ipynb

@@ -0,0 +1,124 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "7eea0d4d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Num GPUs Available:  2\n"
+     ]
+    }
+   ],
+   "source": [
+    "import tensorflow as tf\n",
+    "print(\"Num GPUs Available: \", len(tf.config.list_physical_devices('GPU')))\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "33d18ebd",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2 Physical GPU, 3 Logical GPUs\n"
+     ]
+    }
+   ],
+   "source": [
+    "gpus = tf.config.list_physical_devices('GPU')\n",
+    "if gpus:\n",
+    "  # Create 2 virtual GPUs with 1GB memory each\n",
+    "  try:\n",
+    "    tf.config.set_logical_device_configuration(\n",
+    "        gpus[0],\n",
+    "        [tf.config.LogicalDeviceConfiguration(memory_limit=1024),\n",
+    "         tf.config.LogicalDeviceConfiguration(memory_limit=1024)])\n",
+    "    logical_gpus = tf.config.list_logical_devices('GPU')\n",
+    "    print(len(gpus), \"Physical GPU,\", len(logical_gpus), \"Logical GPUs\")\n",
+    "  except RuntimeError as e:\n",
+    "    # Virtual devices must be set before GPUs have been initialized\n",
+    "    print(e)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "2b9ca96e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "tf.Tensor(\n",
+      "[[22. 28.]\n",
+      " [49. 64.]], shape=(2, 2), dtype=float32)\n"
+     ]
+    }
+   ],
+   "source": [
+    "tf.debugging.set_log_device_placement(True)\n",
+    "\n",
+    "a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])\n",
+    "b = tf.constant([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]])\n",
+    "\n",
+    "# Run on the GPU\n",
+    "c = tf.matmul(a, b)\n",
+    "print(c)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from keras.models import load_model\n",
+    "\n",
+    "# returns a compiled model\n",
+    "# identical to the previous one\n",
+    "model = load_model('Model_1.h5')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model.predict_generator()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

torch_training_version.ipynb

@@ -0,0 +1,209 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "a43c3ccb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch\n",
+    "import torchvision.models as models\n",
+    "import pandas as pd\n",
+    "from torch.utils.data import Dataset, DataLoader\n",
+    "from torchvision import transforms, utils\n",
+    "from matplotlib import pyplot as plt\n",
+    "from matplotlib.image import imread\n",
+    "import pandas as pd\n",
+    "from collections import Counter\n",
+    "import json\n",
+    "import os\n",
+    "import re\n",
+    "import tempfile\n",
+    "from os.path import exists\n",
+    "from PIL import ImageFile\n",
+    "import sklearn as sk\n",
+    "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n",
+    "import image_faults"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "6c7577a6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "2"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "torch.cuda.device_count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "c7e9b947",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "resnet18 = models.resnet18(pretrained=True)\n",
+    "vgg16 = models.vgg16(pretrained=True)\n",
+    "inception = models.inception_v3(pretrained=True)\n",
+    "resnext50_32x4d = models.resnext50_32x4d(pretrained=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "eabc61b2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class Shoes(Dataset):\n",
+    "    def __init__(self, csvfile, root_dir, transform=None):\n",
+    "        self.shoes_df = pd.read_csv(csvfile)\n",
+    "        self.root_dir = root_dir\n",
+    "        self.transform = transform\n",
+    "    \n",
+    "    def __getitem__(self, index):\n",
+    "        self.shoes_df.iloc[index]\n",
+    "        \n",
+    "    \n",
+    "    def __getitem__(self, idx):\n",
+    "        if torch.is_tensor(idx):\n",
+    "            idx = idx.tolist()\n",
+    "\n",
+    "        img_name = os.path.join(self.root_dir,\n",
+    "                                self.data.iloc[idx, 0])\n",
+    "        image = io.imread(img_name)\n",
+    "        data = self.data.iloc[idx, 1:]\n",
+    "        data = np.array([data])\n",
+    "        data = data.astype('float').reshape(-1, 2)\n",
+    "        sample = {'image': image, 'landmarks': data}\n",
+    "\n",
+    "        if self.transform:\n",
+    "            sample = self.transform(sample)\n",
+    "\n",
+    "        return sample\n",
+    "        \n",
+    "    def __len__(self):\n",
+    "        return len(self.data)\n",
+    "        "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "a0fc66b0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "something = pd.read_csv('expanded_class.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "ed2aceeb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def dict_pics_jup():\n",
+    "    '''\n",
+    "    {source:target} dict used to replace source urls with image location as input\n",
+    "    '''\n",
+    "    target_dir = os.getcwd() + os.sep + \"training_images\"\n",
+    "    with open('temp_pics_source_list.txt') as f:\n",
+    "        temp_pics_source_list = json.load(f)\n",
+    "        \n",
+    "    dict_pics = {}\n",
+    "    for k in temp_pics_source_list:\n",
+    "        try:      \n",
+    "             patt_1 = re.search(r'[^/]+(?=/\\$_|.(\\.jpg|\\.jpeg|\\.png))', k, re.IGNORECASE)\n",
+    "             patt_2 = re.search(r'(\\.jpg|\\.jpeg|\\.png)', k, re.IGNORECASE)\n",
+    "             if patt_1 and patt_2 is not None:\n",
+    "                 tag = patt_1.group() + patt_2.group().lower()\n",
+    "                 file_name = target_dir + os.sep + tag\n",
+    "                 dict_pics.update({k:file_name})\n",
+    "        except TypeError:\n",
+    "            print(k)\n",
+    "    print(\"{source:target} dictionary created @ \" + target_dir)\n",
+    "    return dict_pics"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "0095fa33",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def cleanup():\n",
+    "    with open('women_cat_list.txt') as f:\n",
+    "        women_cats = json.load(f)\n",
+    "    with open('men_cat_list.txt') as f:\n",
+    "        men_cats = json.load(f)\n",
+    "\n",
+    "    with open('temp_pics_source_list.txt') as f:\n",
+    "        tempics = json.load(f)\n",
+    "    # list of image urls that did not get named properly which will be removed from the dataframe\n",
+    "    drop_row_vals = []\n",
+    "    for pic in tempics:\n",
+    "        try:\n",
+    "            dict_pics[pic]\n",
+    "        except KeyError:\n",
+    "            drop_row_vals.append(pic)\n",
+    "\n",
+    "    df['PrimaryCategoryID'] = df['PrimaryCategoryID'].astype(str) # pandas thinks ids are ints\n",
+    "    df = df[df.PictureURL.isin(drop_row_vals)==False] # remove improperly named image files\n",
+    "    df = df[df.PrimaryCategoryID.isin(men_cats)==False] # removes rows of womens categories\n",
+    "\n",
+    "    blah = pd.Series(df.PictureURL)\n",
+    "    df = df.drop(labels=['PictureURL'], axis=1)\n",
+    "\n",
+    "    blah = blah.apply(lambda x: dict_pics[x])\n",
+    "    df = pd.concat([blah, df],axis=1)\n",
+    "    df = df.groupby('PrimaryCategoryID').filter(lambda x: len(x)>25) # removes cat outliers\n",
+    "    return df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "edd196dc",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

try.py

@@ -0,0 +1,37 @@
+import ebaysdk
+import json
+import requests
+import concurrent.futures
+import config as cfg
+from ebaysdk.shopping import Connection as Shopping
+from ebaysdk.trading import Connection as Trading
+sapi = Shopping(config_file = 'ebay.yaml')
+tapi = Trading(config_file='ebay.yaml')
+
+def get_cat_specs(cat):
+
+    response = tapi.execute('GetCategorySpecifics',
+            {'CategoryID':cat})
+    cat_spacs =[name['Name'] for name in response.dict()['Recommendations']['NameRecommendation']]
+
+    return cat_spacs
+
+with open('cat_list.txt') as f:
+    cat_list = json.load(f)
+
+def threadd_cat_spacs():
+
+    cat_spacs = []
+
+    with concurrent.futures.ThreadPoolExecutor() as executor:
+        for future in executor.map(get_cat_specs, cat_list):
+            cat_spacs.extend(future)
+
+    cat_spacs = list(set(cat_spacs))
+
+    return cat_spacs
+
+if __name__=='__main__':
+    cat_spacs = threadd_cat_spacs()
+    with open('cat_spacs.txt', 'w') as f:
+        json.dump(cat_spacs, f)

update_dataset.py

@@ -0,0 +1,13 @@
+'''
+Update dataset; instantiates FindingApi and makes call to eBay's Finding Api
+using the findItemsByCategory service. Updates the master_ids list and raw_data.
+'''
+import ebay_api
+
+# Make call to ebay Finding service and return list of twenty_id strings
+finding = ebay_api.FindingApi(4) # 4 is URL paramter for used items
+twenty_id_list = finding.get_ids_from_cats()[0]
+
+# renew oauth token and make call to shopping service to get item data and write to local file
+shopping = ebay_api.ShoppingApi()
+data = shopping.conky(twenty_id_list)