colorizer

Author: l2k2

keras-autoencoder/autoencoder.py

@@ -1,7 +1,9 @@
-from keras.layers import Input, Dense, Flatten, Reshape
+from keras.layers import Input, Dense, Flatten, Reshape, UpSampling2D, Conv2D, MaxPooling2D
 from keras.models import Model, Sequential
 
 from keras.datasets import mnist
+from keras.datasets import fashion_mnist
+
 from keras.callbacks import Callback
 import numpy as np
 import wandb
@@ -13,16 +15,28 @@
 config.encoding_dim = 32
 config.epochs = 1000
 
-(x_train, _), (x_test, _) = mnist.load_data()
+(x_train, _), (x_test, _) = fashion_mnist.load_data()
 
 x_train = x_train.astype('float32') / 255.
 x_test = x_test.astype('float32') / 255.
 
 model = Sequential()
-model.add(Flatten(input_shape=(28,28)))
-model.add(Dense(config.encoding_dim, activation='relu'))
-model.add(Dense(28*28, activation='sigmoid'))
+model.add(Reshape((28,28,1),input_shape=(28,28)))
+model.add(Conv2D(4, (3,3), padding="same", activation="relu"))
+model.add(MaxPooling2D(pool_size=(2,2)))
+model.add(Conv2D(8, (3,3), padding="same", activation="relu"))
+model.add(MaxPooling2D(pool_size=(2,2)))
+model.add(Flatten())
+model.add(Dense(10, activation="relu"))
+model.add(Dense(7*7, activation="relu"))
+model.add(Reshape((7,7,1)))
+model.add(Conv2D(8, (3,3), padding="same", activation="relu"))
+model.add(UpSampling2D())
+model.add(Conv2D(4, (3,3), padding="same", activation="relu"))
+model.add(UpSampling2D())
+model.add(Conv2D(1, (3,3), padding="same", activation="sigmoid"))
 model.add(Reshape((28,28)))
+
 model.compile(optimizer='adam', loss='mse')
 
 model.summary()
@@ -41,7 +55,7 @@ def on_epoch_end(self, epoch, logs):
 model.fit(x_train, x_train,
                 epochs=config.epochs,
                 validation_data=(x_test, x_test), 
-          callbacks=[Images(), WandbCallback()])
+                callbacks=[Images(), WandbCallback()])
 
 
 model.save('auto-small.h5')

keras-autoencoder/denoising_autoencoder.py

@@ -7,9 +7,9 @@
 from wandb.keras import WandbCallback
 
 def add_noise(x_train, x_test):
-    noise_factor = 0.5
-    x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape) 
-    x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape) 
+    noise_factor = 1.0
+    x_train_noisy = x_train + np.random.normal(loc=0.0, scale=noise_factor, size=x_train.shape) 
+    x_test_noisy = x_test + np.random.normal(loc=0.0, scale=noise_factor, size=x_test.shape) 
 
     x_train_noisy = np.clip(x_train_noisy, 0., 1.)
     x_test_noisy = np.clip(x_test_noisy, 0., 1.)
@@ -34,7 +34,7 @@ def add_noise(x_train, x_test):
 model.add(Dense(config.encoding_dim, activation='relu'))
 model.add(Dense(784, activation='sigmoid'))
 model.add(Reshape((28,28)))
-model.compile(optimizer='adam', loss='mse')
+model.compile(optimizer='adam', loss='binary_crossentropy')
 
 class Images(Callback):
       def on_epoch_end(self, epoch, logs):

keras-color/wandb/settings

@@ -1,4 +1,4 @@
 [default]
-entity: qualcomm
-project: color-jul27
+entity: mlclass
+project: color-image
 base_url: https://api.wandb.ai

keras-fashion/perceptron-linear.py

@@ -1,7 +1,7 @@
 import numpy
 from keras.datasets import fashion_mnist
 from keras.models import Sequential
-from keras.layers import Dense, Flatten, Dropout
+from keras.layers import Dense, Flatten, Dropout, Conv2D, MaxPooling2D, Reshape
 from keras.utils import np_utils
 import wandb
 from wandb.keras import WandbCallback
@@ -14,6 +14,10 @@
 # load data
 (X_train, y_train), (X_test, y_test) = fashion_mnist.load_data()
 
+X_train = X_train.astype('float32') / 255.
+X_test = X_test.astype('float32') / 255.
+
+
 img_width = X_train.shape[1]
 img_height = X_train.shape[2]
 labels =["T-shirt/top","Trouser","Pullover","Dress",
@@ -27,14 +31,20 @@
 
 # create model
 model=Sequential()
-model.add(Flatten(input_shape=(img_width,img_height)))
-model.add(Dense(num_classes))
-model.compile(loss='mse', optimizer='adam',
+model.add(Reshape((28,28,1), input_shape=(28,28)))
+model.add(Dropout(0.5))
+model.add(Conv2D(32, (3,3), activation='relu'))
+model.add(Dropout(0.5))
+model.add(MaxPooling2D(pool_size=(2,2)))
+model.add(Flatten())
+model.add(Dense(100, activation='relu'))
+model.add(Dropout(0.5))
+model.add(Dense(num_classes, activation='softmax'))
+model.compile(loss='categorical_crossentropy', optimizer='adam',
                 metrics=['accuracy'])
 
 # Fit the model
-model.fit(X_train, y_train, epochs=config.epochs, validation_data=(X_test, y_test),
+model.fit(X_train, y_train, epochs=10, validation_data=(X_test, y_test),
                     callbacks=[WandbCallback(data_type="image", labels=labels)])
 
 
-

keras-imdb/imdb-cnn.py

@@ -1,51 +1,57 @@
 from keras.preprocessing import sequence
 from keras.models import Sequential
 from keras.layers import Dense, Dropout, Activation
-from keras.layers import Embedding
+from keras.layers import Embedding, LSTM
 from keras.layers import Conv1D, Flatten
 from keras.datasets import imdb
 import wandb
 from wandb.keras import WandbCallback
+import imdb
+import numpy as np
+from keras.preprocessing import text
 
 wandb.init()
 config = wandb.config
 
 # set parameters:
-config.max_features = 5000
-config.maxlen = 400
+config.vocab_size = 1000
+config.maxlen = 1000
 config.batch_size = 32
 config.embedding_dims = 50
 config.filters = 250
 config.kernel_size = 3
 config.hidden_dims = 250
-config.epochs = 2
+config.epochs = 10
 
-(x_train, y_train), (x_test, y_test) = imdb.load_data(num_words=config.max_features)
+(X_train, y_train), (X_test, y_test) = imdb.load_imdb()
 
-x_train = sequence.pad_sequences(x_train, maxlen=config.maxlen)
-x_test = sequence.pad_sequences(x_test, maxlen=config.maxlen)
-print('x_train shape:', x_train.shape)
-print('x_test shape:', x_test.shape)
+tokenizer = text.Tokenizer(num_words=config.vocab_size)
+tokenizer.fit_on_texts(X_train)
+X_train = tokenizer.texts_to_matrix(X_train)
+X_test = tokenizer.texts_to_matrix(X_test)
+
+X_train = sequence.pad_sequences(X_train, maxlen=config.maxlen)
+X_test = sequence.pad_sequences(X_test, maxlen=config.maxlen)
 
-print('Build model...')
 model = Sequential()
-model.add(Embedding(config.max_features,
+model.add(Embedding(config.vocab_size,
                     config.embedding_dims,
                     input_length=config.maxlen))
-model.add(Dropout(0.2))
+model.add(Dropout(0.5))
 model.add(Conv1D(config.filters,
                  config.kernel_size,
                  padding='valid',
                  activation='relu'))
 model.add(Flatten())
 model.add(Dense(config.hidden_dims, activation='relu'))
-model.add(Dropout(0.2))
+model.add(Dropout(0.5))
 model.add(Dense(1, activation='sigmoid'))
 
 model.compile(loss='binary_crossentropy',
               optimizer='adam',
               metrics=['accuracy'])
-model.fit(x_train, y_train,
+
+model.fit(X_train, y_train,
           batch_size=config.batch_size,
           epochs=config.epochs,
-          validation_data=(x_test, y_test))
+          validation_data=(X_test, y_test), callbacks=[WandbCallback()])

keras-imdb/imdb-lstm.py

@@ -15,7 +15,7 @@
 
 # set parameters:
 config.vocab_size = 1000
-config.maxlen = 1000
+config.maxlen = 300
 config.batch_size = 32
 config.embedding_dims = 50
 config.filters = 250
@@ -37,9 +37,8 @@
 model.add(Embedding(config.vocab_size,
                     config.embedding_dims,
                     input_length=config.maxlen))
-model.add(LSTM(config.hidden_dims, dropout=0.2, recurrent_dropout=0.2, activation="sigmoid", internal_activation="sigmoid"))
+model.add(LSTM(config.hidden_dims, activation="sigmoid"))
 model.add(Dense(1, activation='sigmoid'))
-
 model.compile(loss='binary_crossentropy',
               optimizer='rmsprop',
               metrics=['accuracy'])

keras-seq2seq/train.py

@@ -1,37 +1,12 @@
 
-# -*- coding: utf-8 -*-
-'''An implementation of sequence to sequence learning for performing addition
-Input: "535+61"
-Output: "596"
-Padding is handled by using a repeated sentinel character (space)
-Input may optionally be reversed, shown to increase performance in many tasks in:
-"Learning to Execute"
-http://arxiv.org/abs/1410.4615
-and
-"Sequence to Sequence Learning with Neural Networks"
-http://papers.nips.cc/paper/5346-sequence-to-sequence-learning-with-neural-networks.pdf
-Theoretically it introduces shorter term dependencies between source and target.
-Two digits reversed:
-+ One layer LSTM (128 HN), 5k training examples = 99% train/test accuracy in 55 epochs
-Three digits reversed:
-+ One layer LSTM (128 HN), 50k training examples = 99% train/test accuracy in 100 epochs
-Four digits reversed:
-+ One layer LSTM (128 HN), 400k training examples = 99% train/test accuracy in 20 epochs
-Five digits reversed:
-+ One layer LSTM (128 HN), 550k training examples = 99% train/test accuracy in 30 epochs
-'''  # noqa
-
-from __future__ import print_function
 from keras.models import Sequential
 from keras import layers
 import numpy as np
-from six.moves import range
 import wandb
 from wandb.keras import WandbCallback
 
 wandb.init()
 
-
 class CharacterTable(object):
     """Given a set of characters:
     + Encode them to a one hot integer representation
@@ -64,10 +39,7 @@ def decode(self, x, calc_argmax=True):
         return ''.join(self.indices_char[x] for x in x)
 
 
-class colors:
-    ok = '\033[92m'
-    fail = '\033[91m'
-    close = '\033[0m'
+
 
 # Parameters for the model and dataset.
 TRAINING_SIZE = 50000
@@ -193,7 +165,7 @@ class colors:
         print('Q', q[::-1] if REVERSE else q, end=' ')
         print('T', correct, end=' ')
         if correct == guess:
-            print(colors.ok + '☑' + colors.close, end=' ')
+            print('☑', end=' ')
         else:
-            print(colors.fail + '☒' + colors.close, end=' ')
+            print('☒', end=' ')
         print(guess)

keras-sign/perceptron.py

@@ -36,6 +36,7 @@
 # create model
 model=Sequential()
 model.add(Flatten(input_shape=(img_width, img_height)))
+model.add(Dense(100, activation='relu'))
 model.add(Dense(num_classes, activation='softmax'))
 model.compile(loss=config.loss, optimizer=config.optimizer,
                 metrics=['accuracy'])