-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathLSTM
516 lines (406 loc) · 19.6 KB
/
LSTM
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
import numpy as np
import matplotlib.pyplot as plt
import keras
import csv
def read_csv(filename):
phrase = []
emoji = []
with open (filename) as csvDataFile:
csvReader = csv.reader(csvDataFile)
for row in csvReader:
phrase.append(row[0])
emoji.append(row[1])
X = np.asarray(phrase)
Y = np.asarray(emoji, dtype=int)
return X, Y
from google.colab import files
uploaded = files.upload()
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable.
Saving train_emoji.csv to train_emoji (2).csv
X_train, Y_train = read_csv('train_emoji.csv')
uploaded = files.upload()
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable.
Saving test_emoji.csv to test_emoji (2).csv
X_test, Y_test = read_csv('test_emoji.csv')
maxLen = len(max(X_train, key=len).split())
maxLen
10
!pip install emoji
Requirement already satisfied: emoji in /usr/local/lib/python3.10/dist-packages (1.6.3)
import emoji
emoji_dictionary = {"0": "\u2764\uFE0F", # :heart: prints a black instead of red heart depending on the font
"1": ":baseball:",
"2": ":smile:",
"3": ":disappointed:",
"4": ":fork_and_knife:"}
def label_to_emoji(label):
return emoji.emojize(emoji_dictionary[str(label)], use_aliases=True)
#language='alias'
!pip install --upgrade emoji==1.6.3
Requirement already satisfied: emoji==1.6.3 in /usr/local/lib/python3.10/dist-packages (1.6.3)
index = 5
print(X_train[index], label_to_emoji(Y_train[index]))
I love you mum ❤️
Y_oh_train = keras.utils.to_categor)ical(Y_train, 5
Y_oh_test = keras.utils.to_categorical(Y_test, 5)
index = 50
print(Y_train[index], "is converted into one hot", Y_oh_train[index])
0 is converted into one hot [1. 0. 0. 0. 0.]
def read_glove_vecs(glove_file):
with open(glove_file, encoding="utf8") as f:
words = set()
word_to_vec_map = {}
for line in f:
line = line.strip().split()
curr_word = line[0]
words.add(curr_word)
word_to_vec_map[curr_word] = np.array(line[1:], dtype=np.float64)
i = 1
words_to_index = {}
index_to_words = {}
for w in sorted(words):
words_to_index[w] = i
index_to_words[i] = w
i = i + 1
return words_to_index, index_to_words, word_to_vec_map
uploaded = files.upload()
Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable.
Saving glove.6B.50d.txt to glove.6B.50d.txt
word_to_index, index_to_word, word_to_vec_map = read_glove_vecs('glove.6B.50d.txt')
word = "fatemeh"
index = 110011
print("the index of", word, "in the vocabulary is", word_to_index[word])
print("the", str(index) + "th word in the vocabulary is", index_to_word[index])
the index of fatemeh in the vocabulary is 145572
the 110011th word in the vocabulary is cotherstone
word_to_vec_map["fatemeh"]
array([-0.28103 , 0.70142 , 0.95715 , -0.67746 , 0.04095 , 2.1452 ,
1.0724 , -0.24133 , 0.078944, -0.3879 , 0.073104, -0.55003 ,
-0.072175, 0.1631 , 0.011715, 0.45821 , -0.73487 , 0.4834 ,
1.9277 , 0.50251 , -0.087143, 0.37257 , 0.30312 , 0.72518 ,
0.72432 , -0.072846, 0.58899 , 0.50311 , -1.4962 , 0.2182 ,
-1.6104 , 0.75147 , 0.2672 , 0.75233 , 0.12278 , -0.11149 ,
0.56118 , 0.57943 , -0.03375 , 0.15395 , 0.18753 , 0.22272 ,
-0.40814 , -0.020224, 0.2712 , -0.85248 , -0.26494 , 0.46335 ,
1.0987 , -0.45224 ])
def sentence_to_avg(sentence, word_to_vec_map):
words = sentence.lower().split()
avg = np.zeros((50,))
for w in words:
avg += word_to_vec_map[w]
avg = avg / len(words)
return avg
avg = sentence_to_avg("Morrocan couscous is my favorite dish", word_to_vec_map)
print("avg = ", avg)
avg = [-0.008005 0.56370833 -0.50427333 0.258865 0.55131103 0.03104983
-0.21013718 0.16893933 -0.09590267 0.141784 -0.15708967 0.18525867
0.6495785 0.38371117 0.21102167 0.11301667 0.02613967 0.26037767
0.05820667 -0.01578167 -0.12078833 -0.02471267 0.4128455 0.5152061
0.38756167 -0.898661 -0.535145 0.33501167 0.68806933 -0.2156265
1.797155 0.10476933 -0.36775333 0.750785 0.10282583 0.348925
-0.27262833 0.66768 -0.10706167 -0.283635 0.59580117 0.28747333
-0.3366635 0.23393817 0.34349183 0.178405 0.1166155 -0.076433
0.1445417 0.09808667]
def softmax(x):
e_x = np.exp(x - np.max(x))
return e_x / e_x.sum()
def predict(X, Y, W, b, word_to_vec_map):
m = X.shape[0]
pred = np.zeros((m, 1))
for j in range(m):
words = X[j].lower().split()
# Average words' vectors
avg = np.zeros((50,))
for w in words:
avg += word_to_vec_map[w]
avg = avg/len(words)
# Forward propagation
Z = np.dot(W, avg) + b
A = softmax(Z)
pred[j] = np.argmax(A)
print("Accuracy: " + str(np.mean((pred[:] == Y.reshape(Y.shape[0],1)[:]))))
return pred
def model(X, Y, word_to_vec_map, learning_rate = 0.01, num_iterations = 401):
np.random.seed(1)
m = Y.shape[0] # number of training examples
n_y = 5 # number of classes
n_h = 50 # dimensions of the GloVe vectors
W = np.random.randn(n_y, n_h) / np.sqrt(n_h)
b = np.zeros((n_y,))
Y_oh = keras.utils.to_categorical(Y, n_y)
for t in range(num_iterations): # Loop over the number of iterations
for i in range(m): # Loop over the training examples
# Average the word vectors of the words from the i'th training example
avg = sentence_to_avg(X[i], word_to_vec_map)
# Forward propagate the avg through the softmax layer
z = np.dot(W, avg) + b
a = softmax(z)
# Compute cost using the i'th training label's one hot representation and "A" (the output of the softmax)
cost = -np.sum(Y_oh[i] * np.log(a))
# Compute gradients
dz = a - Y_oh[i]
dW = np.dot(dz.reshape(n_y,1), avg.reshape(1, n_h))
db = dz
# Update parameters with Stochastic Gradient Descent
W = W - learning_rate * dW
b = b - learning_rate * db
if t % 100 == 0:
print("Epoch: " + str(t) + " --- cost = " + str(cost))
pred = predict(X, Y, W, b, word_to_vec_map)
return pred, W, b
pred, W, b = model(X_train, Y_train, word_to_vec_map)
Epoch: 0 --- cost = 1.952049881281007
Accuracy: 0.3484848484848485
Epoch: 100 --- cost = 0.07971818726014807
Accuracy: 0.9318181818181818
Epoch: 200 --- cost = 0.04456369243681402
Accuracy: 0.9545454545454546
Epoch: 300 --- cost = 0.03432267378786058
Accuracy: 0.9696969696969697
Epoch: 400 --- cost = 0.02906976783312465
Accuracy: 0.9772727272727273
print("Training set:")
pred_train = predict(X_train, Y_train, W, b, word_to_vec_map)
print('Test set:')
pred_test = predict(X_test, Y_test, W, b, word_to_vec_map)
Training set:
Accuracy: 0.9772727272727273
Test set:
Accuracy: 0.625
def print_predictions(X, pred):
print()
for i in range(X.shape[0]):
print(X[i], label_to_emoji(int(pred[i])))
X_my_sentences = np.array(["i adore you", "i love you", "funny lol", "lets play with a ball", "food is ready", "not feeling happy"])
Y_my_labels = np.array([[0], [0], [2], [1], [4],[3]])
pred = predict(X_my_sentences, Y_my_labels , W, b, word_to_vec_map)
print_predictions(X_my_sentences, pred)
Accuracy: 0.8333333333333334
i adore you ❤️
i love you ❤️
funny lol 😄
lets play with a ball ⚾
food is ready 🍴
not feeling happy 😄
<ipython-input-25-9d81f9387e30>:4: DeprecationWarning: Conversion of an array with ndim > 0 to a scalar is deprecated, and will error in future. Ensure you extract a single element from your array before performing this operation. (Deprecated NumPy 1.25.)
print(X[i], label_to_emoji(int(pred[i])))
import numpy as np
np.random.seed(0)
from keras.models import Model
from keras.layers import Dense, Input, Dropout, LSTM, Activation
#from keras.layers.embeddings import Embedding
from keras.preprocessing import sequence
from keras.initializers import glorot_uniform
np.random.seed(1)
def sentences_to_indices(X, word_to_index, max_len):
m = X.shape[0] # number of training examples
# Initialize X_indices as a numpy matrix of zeros and the correct shape (≈ 1 line)
X_indices = np.zeros((m, max_len))
for i in range(m): # loop over training examples
# Convert the ith training sentence in lower case and split is into words. You should get a list of words.
sentence_words =X[i].lower().split()
# Loop over the words of sentence_words
for j, w in enumerate(sentence_words):
# Set the (i,j)th entry of X_indices to the index of the correct word.
X_indices[i, j] = word_to_index[w]
return X_indices
X1 = np.array(["funny lol", "lets play baseball", "food is ready for you"])
X1_indices = sentences_to_indices(X1,word_to_index, max_len = 5)
print("X1 =", X1)
print("X1_indices =", X1_indices)
X1 = ['funny lol' 'lets play baseball' 'food is ready for you']
X1_indices = [[155345. 225122. 0. 0. 0.]
[220930. 286375. 69714. 0. 0.]
[151204. 192973. 302254. 151349. 394475.]]
from keras.layers import Embedding
def pretrained_embedding_layer(word_to_vec_map, word_to_index):
vocab_len = len(word_to_index) + 1
emb_dim = word_to_vec_map["cucumber"].shape[0]
emb_matrix = np.zeros((vocab_len, emb_dim))
for word, index in word_to_index.items():
emb_matrix[index, :] = word_to_vec_map[word]
embedding_layer = Embedding(vocab_len, emb_dim, trainable = False)
embedding_layer.build((None,))
embedding_layer.set_weights([emb_matrix])
return embedding_layer
from keras.layers import Input
from keras.layers import LSTM
from keras.layers import Dense, Dropout
from keras.models import Model
def Emojify_V2(input_shape, word_to_vec_map, word_to_index):
sentence_indices = Input(input_shape, dtype = np.int32)
embedding_layer = pretrained_embedding_layer(word_to_vec_map, word_to_index)
embeddings = embedding_layer(sentence_indices)
X = LSTM(128, return_sequences=True)(embeddings)
X = Dropout(0.5)(X)
X = LSTM(128)(X)
X = Dropout(0.5)(X)
X = Dense(5, activation = 'softmax')(X)
model = Model(sentence_indices, X)
return model
model = Emojify_V2((maxLen,), word_to_vec_map, word_to_index)
model.summary()
Model: "model"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
input_1 (InputLayer) [(None, 10)] 0
embedding (Embedding) (None, 10, 50) 20000050
lstm (LSTM) (None, 10, 128) 91648
dropout (Dropout) (None, 10, 128) 0
lstm_1 (LSTM) (None, 128) 131584
dropout_1 (Dropout) (None, 128) 0
dense (Dense) (None, 5) 645
=================================================================
Total params: 20223927 (77.15 MB)
Trainable params: 223877 (874.52 KB)
Non-trainable params: 20000050 (76.29 MB)
_________________________________________________________________
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
X_train_indices = sentences_to_indices(X_train, word_to_index, maxLen)
Y_train_oh = keras.utils.to_categorical(Y_train, 5)
model.fit(X_train_indices, Y_train_oh, epochs = 50, batch_size = 32, shuffle=True)
Epoch 1/50
5/5 [==============================] - 6s 16ms/step - loss: 1.5968 - accuracy: 0.2652
Epoch 2/50
5/5 [==============================] - 0s 7ms/step - loss: 1.5005 - accuracy: 0.3030
Epoch 3/50
5/5 [==============================] - 0s 7ms/step - loss: 1.4649 - accuracy: 0.3409
Epoch 4/50
5/5 [==============================] - 0s 8ms/step - loss: 1.3911 - accuracy: 0.4848
Epoch 5/50
5/5 [==============================] - 0s 8ms/step - loss: 1.2764 - accuracy: 0.6212
Epoch 6/50
5/5 [==============================] - 0s 7ms/step - loss: 1.1365 - accuracy: 0.5833
Epoch 7/50
5/5 [==============================] - 0s 7ms/step - loss: 1.0371 - accuracy: 0.5909
Epoch 8/50
5/5 [==============================] - 0s 8ms/step - loss: 0.8725 - accuracy: 0.6591
Epoch 9/50
5/5 [==============================] - 0s 9ms/step - loss: 0.8454 - accuracy: 0.6591
Epoch 10/50
5/5 [==============================] - 0s 9ms/step - loss: 0.7202 - accuracy: 0.6818
Epoch 11/50
5/5 [==============================] - 0s 8ms/step - loss: 0.6708 - accuracy: 0.7500
Epoch 12/50
5/5 [==============================] - 0s 7ms/step - loss: 0.6162 - accuracy: 0.7576
Epoch 13/50
5/5 [==============================] - 0s 7ms/step - loss: 0.8601 - accuracy: 0.6591
Epoch 14/50
5/5 [==============================] - 0s 7ms/step - loss: 0.4961 - accuracy: 0.8409
Epoch 15/50
5/5 [==============================] - 0s 7ms/step - loss: 0.4765 - accuracy: 0.8182
Epoch 16/50
5/5 [==============================] - 0s 7ms/step - loss: 0.4495 - accuracy: 0.8788
Epoch 17/50
5/5 [==============================] - 0s 7ms/step - loss: 0.3787 - accuracy: 0.8788
Epoch 18/50
5/5 [==============================] - 0s 7ms/step - loss: 0.3530 - accuracy: 0.8712
Epoch 19/50
5/5 [==============================] - 0s 7ms/step - loss: 0.3137 - accuracy: 0.8939
Epoch 20/50
5/5 [==============================] - 0s 7ms/step - loss: 0.2268 - accuracy: 0.9242
Epoch 21/50
5/5 [==============================] - 0s 8ms/step - loss: 0.2238 - accuracy: 0.9091
Epoch 22/50
5/5 [==============================] - 0s 7ms/step - loss: 0.2539 - accuracy: 0.9091
Epoch 23/50
5/5 [==============================] - 0s 8ms/step - loss: 0.1893 - accuracy: 0.9470
Epoch 24/50
5/5 [==============================] - 0s 8ms/step - loss: 0.2254 - accuracy: 0.9015
Epoch 25/50
5/5 [==============================] - 0s 7ms/step - loss: 0.2032 - accuracy: 0.9318
Epoch 26/50
5/5 [==============================] - 0s 7ms/step - loss: 0.2125 - accuracy: 0.9394
Epoch 27/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1296 - accuracy: 0.9470
Epoch 28/50
5/5 [==============================] - 0s 10ms/step - loss: 0.1272 - accuracy: 0.9470
Epoch 29/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1000 - accuracy: 0.9773
Epoch 30/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0951 - accuracy: 0.9773
Epoch 31/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1077 - accuracy: 0.9773
Epoch 32/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0906 - accuracy: 0.9697
Epoch 33/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0664 - accuracy: 0.9773
Epoch 34/50
5/5 [==============================] - 0s 7ms/step - loss: 0.3837 - accuracy: 0.9242
Epoch 35/50
5/5 [==============================] - 0s 7ms/step - loss: 0.4920 - accuracy: 0.8864
Epoch 36/50
5/5 [==============================] - 0s 7ms/step - loss: 0.8496 - accuracy: 0.7955
Epoch 37/50
5/5 [==============================] - 0s 8ms/step - loss: 0.4390 - accuracy: 0.8409
Epoch 38/50
5/5 [==============================] - 0s 8ms/step - loss: 0.3310 - accuracy: 0.9091
Epoch 39/50
5/5 [==============================] - 0s 7ms/step - loss: 0.3147 - accuracy: 0.8939
Epoch 40/50
5/5 [==============================] - 0s 7ms/step - loss: 0.2953 - accuracy: 0.9015
Epoch 41/50
5/5 [==============================] - 0s 8ms/step - loss: 0.2288 - accuracy: 0.9167
Epoch 42/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1929 - accuracy: 0.9545
Epoch 43/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1396 - accuracy: 0.9621
Epoch 44/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1445 - accuracy: 0.9697
Epoch 45/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1601 - accuracy: 0.9470
Epoch 46/50
5/5 [==============================] - 0s 7ms/step - loss: 0.1029 - accuracy: 0.9848
Epoch 47/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0904 - accuracy: 0.9848
Epoch 48/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0737 - accuracy: 0.9848
Epoch 49/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0618 - accuracy: 0.9848
Epoch 50/50
5/5 [==============================] - 0s 7ms/step - loss: 0.0467 - accuracy: 0.9924
<keras.src.callbacks.History at 0x7a293433ce80>
X_test_indices = sentences_to_indices(X_test, word_to_index, max_len = maxLen)
Y_test_oh = keras.utils.to_categorical(Y_test, 5)
loss, acc = model.evaluate(X_test_indices, Y_test_oh)
print()
print("Test accuracy = ", acc)
2/2 [==============================] - 1s 9ms/step - loss: 1.1644 - accuracy: 0.6607
Test accuracy = 0.6607142686843872
# This code allows you to see the mislabelled examples
C = 5
y_test_oh = np.eye(C)[Y_test.reshape(-1)]
X_test_indices = sentences_to_indices(X_test, word_to_index, maxLen)
pred = model.predict(X_test_indices)
for i in range(len(X_test)):
x = X_test_indices
num = np.argmax(pred[i])
if(num != Y_test[i]):
print('Expected emoji:'+ label_to_emoji(Y_test[i]) + ' prediction: '+ X_test[i] + label_to_emoji(num).strip())
2/2 [==============================] - 1s 9ms/step
Expected emoji:😄 prediction: he got a raise 😞
Expected emoji:❤️ prediction: he is a good friend 😄
Expected emoji:❤️ prediction: I am upset 😞
Expected emoji:❤️ prediction: We had such a lovely dinner tonight 😄
Expected emoji:😞 prediction: This girl is messing with me ❤️
Expected emoji:😄 prediction: are you serious ha ha 😞
Expected emoji:😄 prediction: you brighten my day ❤️
Expected emoji:😞 prediction: she is a bully ❤️
Expected emoji:😞 prediction: I worked during my birthday 😄
Expected emoji:😄 prediction: enjoy your break 😞
Expected emoji:❤️ prediction: valentine day is near 😄
Expected emoji:😞 prediction: My life is so boring ❤️
Expected emoji:🍴 prediction: I am starving 😞
Expected emoji:😄 prediction: I like your jacket ❤️
Expected emoji:😄 prediction: I want to joke 😞
Expected emoji:😞 prediction: go away ⚾
Expected emoji:😞 prediction: yesterday we lost again ⚾
Expected emoji:❤️ prediction: family is all I have 😞
Expected emoji:😞 prediction: I did not have breakfast 🍴
# Change the sentence below to see your prediction. Make sure all the words are in the Glove embeddings.
x_test = np.array(['you are happy'])
X_test_indices = sentences_to_indices(x_test, word_to_index, maxLen)
print(x_test[0] +' '+ label_to_emoji(np.argmax(model.predict(X_test_indices))))
1/1 [==============================] - 0s 42ms/step
you are happy ❤️