exp-7-text-sequence-generator-lstm

September 25, 2023

1 Sequence to Sequense Prediction of Text

[1]: import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, Embedding

[2]: t_data = 'I am a beautiful person who has been given all the best things in␣
↪life but was not given power to work on them.'

[3]: t_data

[3]: 'I am a beautiful person who has been given all the best things in life but was
not given power to work on them.'

2 Tokenization and Encoding

[4]: tokenizer = tf.keras.preprocessing.text.Tokenizer()
tokenizer.fit_on_texts([t_data])
total_words = len(tokenizer.word_index) + 1

[5]: word_ids = tokenizer.word_index

print(word_ids)

{'given': 1, 'i': 2, 'am': 3, 'a': 4, 'beautiful': 5, 'person': 6, 'who': 7,

'has': 8, 'been': 9, 'all': 10, 'the': 11, 'best': 12, 'things': 13, 'in': 14,
'life': 15, 'but': 16, 'was': 17, 'not': 18, 'power': 19, 'to': 20, 'work': 21,
'on': 22, 'them': 23}

3 Convert the text into sequences for training the model

[6]: input_seq = []
for line in t_data.split("."):
token_list = tokenizer.texts_to_sequences([line])[0]
print(token_list)
print(len(token_list))
for i in range(1, len(token_list)):

1
 num_of_words_per_seq = token_list[:i+1]
input_seq.append(num_of_words_per_seq)
print(input_seq)

[2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 1, 19, 20, 21,
22, 23]
24
[]
0
[[2, 3], [2, 3, 4], [2, 3, 4, 5], [2, 3, 4, 5, 6], [2, 3, 4, 5, 6, 7], [2, 3, 4,
5, 6, 7, 8], [2, 3, 4, 5, 6, 7, 8, 9], [2, 3, 4, 5, 6, 7, 8, 9, 1], [2, 3, 4, 5,
6, 7, 8, 9, 1, 10], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11], [2, 3, 4, 5, 6, 7, 8,
9, 1, 10, 11, 12], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13], [2, 3, 4, 5, 6,
7, 8, 9, 1, 10, 11, 12, 13, 14], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14,
15], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16], [2, 3, 4, 5, 6, 7,
8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11,
12, 13, 14, 15, 16, 17, 18], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15,
16, 17, 18, 1], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18,
1, 19], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 1, 19,
20], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 1, 19, 20,
21], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 1, 19, 20,
21, 22], [2, 3, 4, 5, 6, 7, 8, 9, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 1, 19,
20, 21, 22, 23]]

[7]: print(len(input_seq))

23

[8]: maximum_length_of_seq = max(len(seq) for seq in input_seq)

maximum_length_of_seq

[8]: 24

[9]: input_seq_padded = tf.keras.preprocessing.sequence.pad_sequences(input_seq,␣

↪maxlen=maximum_length_of_seq, padding='pre')

print(input_seq_padded)

[[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11]
[ 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12]

2
 [ 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13]
[ 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14]
[ 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15]
[ 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16]
[ 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17]
[ 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18]
[ 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1]
[ 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19]
[ 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20]
[ 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20 21]
[ 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20 21 22]
[ 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20 21 22 23]]

4 Construct the training and testing DATA

[10]: X, y = input_seq_padded[:, :-1], input_seq_padded[:, -1]

[11]: y = tf.keras.utils.to_categorical(y, num_classes=total_words)

[12]: print(X)

[[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10]
[ 0 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11]
[ 0 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12]
[ 0 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13]
[ 0 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14]
[ 0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15]
[ 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16]
[ 0 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17]
[ 0 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18]
[ 0 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1]
[ 0 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19]
[ 0 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20]
[ 0 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20 21]
[ 2 3 4 5 6 7 8 9 1 10 11 12 13 14 15 16 17 18 1 19 20 21 22]]

[13]: print(y)

3
 [[0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]
[0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]]

5 LSTM Model Building using Tensorflow Library

[14]: model = Sequential()
model.add(Embedding(total_words, 50,input_length=maximum_length_of_seq-1))
model.add(LSTM(100))
model.add(Dense(total_words, activation='softmax'))

[15]: model.summary()

Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
embedding (Embedding) (None, 23, 50) 1200

lstm (LSTM) (None, 100) 60400

dense (Dense) (None, 24) 2424

=================================================================
Total params: 64,024
Trainable params: 64,024
Non-trainable params: 0

4
 _________________________________________________________________

6 Compile the model

[16]: model.compile(loss='categorical_crossentropy', optimizer='adam',␣
↪metrics=['accuracy'])

7 Training the model

[37]: model.fit(X, y, epochs=99, verbose=1)

Epoch 1/99
1/1 [==============================] - 0s 16ms/step - loss: 0.4417 - accuracy:
0.9565
Epoch 2/99
1/1 [==============================] - 0s 17ms/step - loss: 0.5865 - accuracy:
0.8261
Epoch 3/99
1/1 [==============================] - 0s 16ms/step - loss: 0.6158 - accuracy:
0.7391
Epoch 4/99
1/1 [==============================] - 0s 15ms/step - loss: 0.6683 - accuracy:
0.7391
Epoch 5/99
1/1 [==============================] - 0s 12ms/step - loss: 0.4220 - accuracy:
1.0000
Epoch 6/99
1/1 [==============================] - 0s 12ms/step - loss: 0.6846 - accuracy:
0.5217
Epoch 7/99
1/1 [==============================] - 0s 4ms/step - loss: 0.3843 - accuracy:
1.0000
Epoch 8/99
1/1 [==============================] - 0s 6ms/step - loss: 0.5402 - accuracy:
0.7826
Epoch 9/99
1/1 [==============================] - 0s 6ms/step - loss: 0.4710 - accuracy:
0.9130
Epoch 10/99
1/1 [==============================] - 0s 16ms/step - loss: 0.3784 - accuracy:
1.0000
Epoch 11/99
1/1 [==============================] - 0s 18ms/step - loss: 0.4984 - accuracy:
0.8696
Epoch 12/99
1/1 [==============================] - 0s 15ms/step - loss: 0.4065 - accuracy:

5
0.9565
Epoch 13/99
1/1 [==============================] - 0s 17ms/step - loss: 0.3802 - accuracy:
1.0000
Epoch 14/99
1/1 [==============================] - 0s 16ms/step - loss: 0.4489 - accuracy:
0.9565
Epoch 15/99
1/1 [==============================] - 0s 16ms/step - loss: 0.3820 - accuracy:
1.0000
Epoch 16/99
1/1 [==============================] - 0s 15ms/step - loss: 0.3602 - accuracy:
1.0000
Epoch 17/99
1/1 [==============================] - 0s 14ms/step - loss: 0.3982 - accuracy:
0.9565
Epoch 18/99
1/1 [==============================] - 0s 14ms/step - loss: 0.3717 - accuracy:
1.0000
Epoch 19/99
1/1 [==============================] - 0s 13ms/step - loss: 0.3408 - accuracy:
1.0000
Epoch 20/99
1/1 [==============================] - 0s 1ms/step - loss: 0.3602 - accuracy:
1.0000
Epoch 21/99
1/1 [==============================] - 0s 4ms/step - loss: 0.3641 - accuracy:
1.0000
Epoch 22/99
1/1 [==============================] - 0s 18ms/step - loss: 0.3372 - accuracy:
1.0000
Epoch 23/99
1/1 [==============================] - 0s 18ms/step - loss: 0.3319 - accuracy:
1.0000
Epoch 24/99
1/1 [==============================] - 0s 16ms/step - loss: 0.3473 - accuracy:
1.0000
Epoch 25/99
1/1 [==============================] - 0s 15ms/step - loss: 0.3393 - accuracy:
1.0000
Epoch 26/99
1/1 [==============================] - 0s 17ms/step - loss: 0.3222 - accuracy:
1.0000
Epoch 27/99
1/1 [==============================] - 0s 16ms/step - loss: 0.3257 - accuracy:
1.0000
Epoch 28/99
1/1 [==============================] - 0s 15ms/step - loss: 0.3330 - accuracy:

6
1.0000
Epoch 29/99
1/1 [==============================] - 0s 13ms/step - loss: 0.3230 - accuracy:
1.0000
Epoch 30/99
1/1 [==============================] - 0s 12ms/step - loss: 0.3125 - accuracy:
1.0000
Epoch 31/99
1/1 [==============================] - 0s 3ms/step - loss: 0.3156 - accuracy:
1.0000
Epoch 32/99
1/1 [==============================] - 0s 7ms/step - loss: 0.3179 - accuracy:
1.0000
Epoch 33/99
1/1 [==============================] - 0s 5ms/step - loss: 0.3089 - accuracy:
1.0000
Epoch 34/99
1/1 [==============================] - 0s 20ms/step - loss: 0.3028 - accuracy:
1.0000
Epoch 35/99
1/1 [==============================] - 0s 16ms/step - loss: 0.3048 - accuracy:
1.0000
Epoch 36/99
1/1 [==============================] - 0s 17ms/step - loss: 0.3044 - accuracy:
1.0000
Epoch 37/99
1/1 [==============================] - 0s 14ms/step - loss: 0.2979 - accuracy:
1.0000
Epoch 38/99
1/1 [==============================] - 0s 19ms/step - loss: 0.2938 - accuracy:
1.0000
Epoch 39/99
1/1 [==============================] - 0s 16ms/step - loss: 0.2947 - accuracy:
1.0000
Epoch 40/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2935 - accuracy:
1.0000
Epoch 41/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2885 - accuracy:
1.0000
Epoch 42/99
1/1 [==============================] - 0s 15ms/step - loss: 0.2859 - accuracy:
1.0000
Epoch 43/99
1/1 [==============================] - 0s 16ms/step - loss: 0.2860 - accuracy:
1.0000
Epoch 44/99
1/1 [==============================] - 0s 18ms/step - loss: 0.2842 - accuracy:

7
1.0000
Epoch 45/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2804 - accuracy:
1.0000
Epoch 46/99
1/1 [==============================] - 0s 7ms/step - loss: 0.2782 - accuracy:
1.0000
Epoch 47/99
1/1 [==============================] - 0s 5ms/step - loss: 0.2777 - accuracy:
1.0000
Epoch 48/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2756 - accuracy:
1.0000
Epoch 49/99
1/1 [==============================] - 0s 2ms/step - loss: 0.2723 - accuracy:
1.0000
Epoch 50/99
1/1 [==============================] - 0s 8ms/step - loss: 0.2705 - accuracy:
1.0000
Epoch 51/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2695 - accuracy:
1.0000
Epoch 52/99
1/1 [==============================] - 0s 8ms/step - loss: 0.2672 - accuracy:
1.0000
Epoch 53/99
1/1 [==============================] - 0s 8ms/step - loss: 0.2644 - accuracy:
1.0000
Epoch 54/99
1/1 [==============================] - 0s 7ms/step - loss: 0.2628 - accuracy:
1.0000
Epoch 55/99
1/1 [==============================] - 0s 9ms/step - loss: 0.2616 - accuracy:
1.0000
Epoch 56/99
1/1 [==============================] - 0s 7ms/step - loss: 0.2595 - accuracy:
1.0000
Epoch 57/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2572 - accuracy:
1.0000
Epoch 58/99
1/1 [==============================] - 0s 13ms/step - loss: 0.2557 - accuracy:
1.0000
Epoch 59/99
1/1 [==============================] - 0s 15ms/step - loss: 0.2542 - accuracy:
1.0000
Epoch 60/99
1/1 [==============================] - 0s 12ms/step - loss: 0.2522 - accuracy:

8
1.0000
Epoch 61/99
1/1 [==============================] - 0s 16ms/step - loss: 0.2502 - accuracy:
1.0000
Epoch 62/99
1/1 [==============================] - 0s 13ms/step - loss: 0.2487 - accuracy:
1.0000
Epoch 63/99
1/1 [==============================] - 0s 1ms/step - loss: 0.2471 - accuracy:
1.0000
Epoch 64/99
1/1 [==============================] - 0s 5ms/step - loss: 0.2451 - accuracy:
1.0000
Epoch 65/99
1/1 [==============================] - 0s 5ms/step - loss: 0.2433 - accuracy:
1.0000
Epoch 66/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2418 - accuracy:
1.0000
Epoch 67/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2401 - accuracy:
1.0000
Epoch 68/99
1/1 [==============================] - 0s 13ms/step - loss: 0.2382 - accuracy:
1.0000
Epoch 69/99
1/1 [==============================] - 0s 2ms/step - loss: 0.2366 - accuracy:
1.0000
Epoch 70/99
1/1 [==============================] - 0s 5ms/step - loss: 0.2351 - accuracy:
1.0000
Epoch 71/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2333 - accuracy:
1.0000
Epoch 72/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2316 - accuracy:
1.0000
Epoch 73/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2301 - accuracy:
1.0000
Epoch 74/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2284 - accuracy:
1.0000
Epoch 75/99
1/1 [==============================] - 0s 15ms/step - loss: 0.2267 - accuracy:
1.0000
Epoch 76/99
1/1 [==============================] - 0s 14ms/step - loss: 0.2251 - accuracy:

9
1.0000
Epoch 77/99
1/1 [==============================] - 0s 12ms/step - loss: 0.2236 - accuracy:
1.0000
Epoch 78/99
1/1 [==============================] - 0s 12ms/step - loss: 0.2219 - accuracy:
1.0000
Epoch 79/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2203 - accuracy:
1.0000
Epoch 80/99
1/1 [==============================] - 0s 6ms/step - loss: 0.2188 - accuracy:
1.0000
Epoch 81/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2172 - accuracy:
1.0000
Epoch 82/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2156 - accuracy:
1.0000
Epoch 83/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2140 - accuracy:
1.0000
Epoch 84/99
1/1 [==============================] - 0s 12ms/step - loss: 0.2125 - accuracy:
1.0000
Epoch 85/99
1/1 [==============================] - 0s 13ms/step - loss: 0.2109 - accuracy:
1.0000
Epoch 86/99
1/1 [==============================] - 0s 15ms/step - loss: 0.2094 - accuracy:
1.0000
Epoch 87/99
1/1 [==============================] - 0s 0s/step - loss: 0.2079 - accuracy:
1.0000
Epoch 88/99
1/1 [==============================] - 0s 5ms/step - loss: 0.2063 - accuracy:
1.0000
Epoch 89/99
1/1 [==============================] - 0s 16ms/step - loss: 0.2048 - accuracy:
1.0000
Epoch 90/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2033 - accuracy:
1.0000
Epoch 91/99
1/1 [==============================] - 0s 17ms/step - loss: 0.2018 - accuracy:
1.0000
Epoch 92/99
1/1 [==============================] - 0s 16ms/step - loss: 0.2002 - accuracy:

10
 1.0000
Epoch 93/99
1/1 [==============================] - 0s 16ms/step - loss: 0.1988 - accuracy:
1.0000
Epoch 94/99
1/1 [==============================] - 0s 14ms/step - loss: 0.1973 - accuracy:
1.0000
Epoch 95/99
1/1 [==============================] - 0s 866us/step - loss: 0.1958 - accuracy:
1.0000
Epoch 96/99
1/1 [==============================] - 0s 4ms/step - loss: 0.1943 - accuracy:
1.0000
Epoch 97/99
1/1 [==============================] - 0s 18ms/step - loss: 0.1928 - accuracy:
1.0000
Epoch 98/99
1/1 [==============================] - 0s 16ms/step - loss: 0.1914 - accuracy:
1.0000
Epoch 99/99
1/1 [==============================] - 0s 17ms/step - loss: 0.1899 - accuracy:
1.0000

[37]: <keras.callbacks.History at 0x15625b2dcd0>

8 Now Lets check how the model generates Text

[38]: your_seed_text = "You:I"
next_words = 23

[39]: for _ in range(next_words):

token_list = tokenizer.texts_to_sequences([your_seed_text])[0]
token_list = tf.keras.preprocessing.sequence.pad_sequences([token_list],␣
↪maxlen=maximum_length_of_seq-1, padding='pre')

predicted = model.predict(token_list, verbose=0)

predicted_word_index = tf.argmax(predicted, axis=-1)
output_word = ""
for word, index in tokenizer.word_index.items():
if index == predicted_word_index:
output_word = word
break
your_seed_text += " " + output_word
print(f"Model: {your_seed_text.replace('You:', '')}")

Model: I am a beautiful person who has been given all the best things in life
but was not given power to work on them

11
[40]: print(your_seed_text)

You:I am a beautiful person who has been given all the best things in life but
was not given power to work on them

[20]:

12