Merge branch 'master' of gitlab.cl.uni-heidelberg.de:jmrs/caption (1685f7bf) · Commits · Softwareprojekt 2018 - Jason Marinco Rudolf Samuel / O Caption My Caption

caption-lib/.vscode/launch.json

0 → 100644

+123 −0

Original line number	Diff line number	Diff line
		{
		// Use IntelliSense to learn about possible attributes.
		// Hover to view descriptions of existing attributes.
		// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
		"version": "0.2.0",
		"configurations": [

		{
		"name": "Python: Current File",
		"type": "python",
		"request": "launch",
		"program": "${file}"
		},
		{
		"name": "Python: START.py",
		"type": "python",
		"request": "launch",
		"program": "${workspaceFolder}/start.py"
		},
		{
		"name": "Python: Attach",
		"type": "python",
		"request": "attach",
		"localRoot": "${workspaceFolder}",
		"remoteRoot": "${workspaceFolder}",
		"port": 3000,
		"secret": "my_secret",
		"host": "localhost"
		},
		{
		"name": "Python: Terminal (integrated)",
		"type": "python",
		"request": "launch",
		"program": "${file}",
		"console": "integratedTerminal"
		},
		{
		"name": "Python: Terminal (external)",
		"type": "python",
		"request": "launch",
		"program": "${file}",
		"console": "externalTerminal"
		},
		{
		"name": "Python: Django",
		"type": "python",
		"request": "launch",
		"program": "${workspaceFolder}/manage.py",
		"args": [
		"runserver",
		"--noreload",
		"--nothreading"
		],
		"debugOptions": [
		"RedirectOutput",
		"Django"
		]
		},
		{
		"name": "Python: Flask (0.11.x or later)",
		"type": "python",
		"request": "launch",
		"module": "flask",
		"env": {
		"FLASK_APP": "${workspaceFolder}/app.py"
		},
		"args": [
		"run",
		"--no-debugger",
		"--no-reload"
		]
		},
		{
		"name": "Python: Module",
		"type": "python",
		"request": "launch",
		"module": "module.name"
		},
		{
		"name": "Python: Pyramid",
		"type": "python",
		"request": "launch",
		"args": [
		"${workspaceFolder}/development.ini"
		],
		"debugOptions": [
		"RedirectOutput",
		"Pyramid"
		]
		},
		{
		"name": "Python: Watson",
		"type": "python",
		"request": "launch",
		"program": "${workspaceFolder}/console.py",
		"args": [
		"dev",
		"runserver",
		"--noreload=True"
		]
		},
		{
		"name": "Python: All debug Options",
		"type": "python",
		"request": "launch",
		"pythonPath": "${config:python.pythonPath}",
		"program": "${file}",
		"module": "module.name",
		"env": {
		"VAR1": "1",
		"VAR2": "2"
		},
		"envFile": "${workspaceFolder}/.env",
		"args": [
		"arg1",
		"arg2"
		],
		"debugOptions": [
		"RedirectOutput"
		]
		}
		]
		}
		No newline at end of file

caption-lib/lstm/Readers/init.py

+2 −0

Original line number	Diff line number	Diff line
		# reader init
		from .basic_text_reader import basic_text_reader
		from .basic_text_reader import get_words_as_int
		from .basic_text_reader import get_integers_as_words
		from .basic_masa_reader import basic_masa_reader
		from .helper import get_input_data
		from .helper import get_vocabulary

caption-lib/lstm/Readers/basic_text_reader.py

+35 −2

Original line number	Diff line number	Diff line
		@@ -2,10 +2,11 @@

		from lstm.configuration import current as config
		import collections
		import tensorflow as tf
		import json as jsonlib
		import os
		from datetime import datetime
		from lstm.Readers import helper
		import tensorflow as tf

		parent_folder_path = os.path.dirname(__file__) + "/.."
		inputPath = parent_folder_path + config.path.input_folder_path
		@@ -49,7 +50,11 @@ def read_input(filename):

		'''
		Create sequences and save them
		'''
		example:
		batches = [
		[['You', 'can', 'see', 'a', 'cat', '<EOS>'], ['You', 'can', 'see', 'a', 'dog', 'on', 'a', 'table', '<EOS>']]
		[['You', 'can', 'see', 'a', 'girafe', 'on', 'a', 'table', '<EOS>'], ['You', 'see', 'a', 'hotdog', '<EOS>']]
		]'''
		def create_sequences(filename):
		with open(inputPath + filename,"r") as file:
		text=file.read()
		@@ -114,3 +119,31 @@ def save_vocabulary(data, filename):
		if(config.reader.verbose):
		print("Reader:: Saved vocabulary at:", filename)
		print("[" + str(datetime.now().time()) + "]" + "Reader:: ...Finished saving vocabulary!")

		'''
		revert text input list [x, y, z]
		to int list with vocab dict [23, 2342, 123]
		'''
		def get_words_as_int(sequence_list, filename):
		# print("Words to Integers")
		# print("input: ", sequence_list)
		vocab = helper.get_vocabulary(filename)
		# print("vocab: ", vocab)
		sequence = [vocab[word] for word in sequence_list if word in vocab]
		# print("output: ", sequence)
		return sequence

		'''
		revert int list of words [23,2342,123]
		to string list with vocab dict [x, y, z]
		'''
		def get_integers_as_words(sequence_list, filename):
		# print("Integers to Words")
		# print("input: ", sequence_list)
		vocab = helper.get_vocabulary(filename)
		# print("vocab: ", vocab)
		id2word = {i: c for (c, i) in vocab.items()}
		# print("vocab rewersed: ", id2word)
		sequence = [id2word[integer_id] for integer_id in sequence_list if integer_id in id2word]
		# print("output: ", sequence)
		return sequence

caption-lib/lstm/configuration.py

+8 −5

Original line number	Diff line number	Diff line
		@@ -32,17 +32,19 @@ class default_validator(object):

		class default_trainer(object):
		train_on = True
		verbose = False
		padding = 0
		end_of_sentence = 1
		vocab_size = 10 # a word size
		input_embedding_size = 20 # length of the charater
		encoder_hidden_units = 20
		decoder_hidden_units = encoder_hidden_units * 2
		max_batches = 300
		max_batches = 2
		batches_in_epoch = 10
		batch_size = 10
		PAD = 0
		EOS = 1
		batch_size = 2
		sequence_size = 9
		PAD = ""
		EOS = '<EOS>'


		'''
		@@ -61,4 +63,5 @@ class current(default_reader, default_validator, default_trainer, default_paths)
		reader.reader_on = False
		reader.verbose = False
		validator.validator_on = False
		trainer.train_on = True
		trainer.train_on = False
		trainer.verbose = True

caption-lib/lstm/helpers.py

+1 −1

Original line number	Diff line number	Diff line
		@@ -24,7 +24,7 @@ def batch(inputs, max_sequence_length=None):
		if max_sequence_length is None:
		max_sequence_length = max(sequence_lengths)

		inputs_batch_major = np.zeros(shape=[batch_size, max_sequence_length], dtype=np.int32) # == PAD
		inputs_batch_major = np.array([[None] * max_sequence_length] * batch_size) # == PAD

		for i, seq in enumerate(inputs):
		for j, element in enumerate(seq):