practical-no12

October 15, 2024

[1]: import numpy as np

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Embedding, LSTM, GRU, Dense
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from tensorflow.keras.utils import to_categorical

WARNING:tensorflow:From C:\Users\shivr\anaconda3\Lib\site-
packages\keras\src\losses.py:2976: The name
tf.losses.sparse_softmax_cross_entropy is deprecated. Please use
tf.compat.v1.losses.sparse_softmax_cross_entropy instead.

[2]: # Example preprocessed dataset

sentences = [
"The bank was flooded with water after the storm.",
"He deposited the money at the bank."
]
labels = ["river_bank", "financial_institution"] # Word senses for "bank"

[3]: # Tokenization and preprocessing

vocab_size = 10000 # Just an example size, adjust as needed
max_length = 10 # Max length of a sentence

# Assuming sentences are tokenized and converted to integers

X_data = np.random.randint(vocab_size, size=(len(sentences), max_length))

# Label encoding for senses

label_encoder = LabelEncoder()
y_data = label_encoder.fit_transform(labels)
y_data = to_categorical(y_data) # One-hot encoding for categorical output

[4]: # Split into train and test sets

X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.
↪2)

# Define the model (LSTM or GRU)

1
 def build_model(rnn_type='LSTM'):
model = Sequential()
model.add(Embedding(input_dim=vocab_size, output_dim=64,␣
↪input_length=max_length))

if rnn_type == 'LSTM':
model.add(LSTM(128, return_sequences=False))
elif rnn_type == 'GRU':
model.add(GRU(128, return_sequences=False))

model.add(Dense(y_data.shape[1], activation='softmax')) # Output layer for␣

↪sense classification

# Compile the model

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

return model

[5]: # Build and train the LSTM model

model = build_model(rnn_type='LSTM')
model.summary()

WARNING:tensorflow:From C:\Users\shivr\anaconda3\Lib\site-
packages\keras\src\backend.py:873: The name tf.get_default_graph is deprecated.
Please use tf.compat.v1.get_default_graph instead.

WARNING:tensorflow:From C:\Users\shivr\anaconda3\Lib\site-
packages\keras\src\optimizers\__init__.py:309: The name tf.train.Optimizer is
deprecated. Please use tf.compat.v1.train.Optimizer instead.

Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
embedding (Embedding) (None, 10, 64) 640000

lstm (LSTM) (None, 128) 98816

dense (Dense) (None, 2) 258

=================================================================
Total params: 739074 (2.82 MB)
Trainable params: 739074 (2.82 MB)
Non-trainable params: 0 (0.00 Byte)
_________________________________________________________________

2
[6]: # Train the model
model.fit(X_train, y_train, epochs=50, batch_size=32)

Epoch 1/50
WARNING:tensorflow:From C:\Users\shivr\anaconda3\Lib\site-
packages\keras\src\utils\tf_utils.py:492: The name tf.ragged.RaggedTensorValue
is deprecated. Please use tf.compat.v1.ragged.RaggedTensorValue instead.

WARNING:tensorflow:From C:\Users\shivr\anaconda3\Lib\site-
packages\keras\src\engine\base_layer_utils.py:384: The name
tf.executing_eagerly_outside_functions is deprecated. Please use
tf.compat.v1.executing_eagerly_outside_functions instead.

1/1 [==============================] - 3s 3s/step - loss: 0.7011 - accuracy:

0.0000e+00
Epoch 2/50
1/1 [==============================] - 0s 14ms/step - loss: 0.6783 - accuracy:
1.0000
Epoch 3/50
1/1 [==============================] - 0s 10ms/step - loss: 0.6556 - accuracy:
1.0000
Epoch 4/50
1/1 [==============================] - 0s 11ms/step - loss: 0.6317 - accuracy:
1.0000
Epoch 5/50
1/1 [==============================] - 0s 11ms/step - loss: 0.6053 - accuracy:
1.0000
Epoch 6/50
1/1 [==============================] - 0s 11ms/step - loss: 0.5753 - accuracy:
1.0000
Epoch 7/50
1/1 [==============================] - 0s 10ms/step - loss: 0.5406 - accuracy:
1.0000
Epoch 8/50
1/1 [==============================] - 0s 10ms/step - loss: 0.5000 - accuracy:
1.0000
Epoch 9/50
1/1 [==============================] - 0s 10ms/step - loss: 0.4523 - accuracy:
1.0000
Epoch 10/50
1/1 [==============================] - 0s 10ms/step - loss: 0.3966 - accuracy:
1.0000
Epoch 11/50
1/1 [==============================] - 0s 10ms/step - loss: 0.3329 - accuracy:
1.0000
Epoch 12/50
1/1 [==============================] - 0s 10ms/step - loss: 0.2620 - accuracy:

3
1.0000
Epoch 13/50
1/1 [==============================] - 0s 11ms/step - loss: 0.1877 - accuracy:
1.0000
Epoch 14/50
1/1 [==============================] - 0s 10ms/step - loss: 0.1169 - accuracy:
1.0000
Epoch 15/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0593 - accuracy:
1.0000
Epoch 16/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0228 - accuracy:
1.0000
Epoch 17/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0063 - accuracy:
1.0000
Epoch 18/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0013 - accuracy:
1.0000
Epoch 19/50
1/1 [==============================] - 0s 9ms/step - loss: 2.1324e-04 -
accuracy: 1.0000
Epoch 20/50
1/1 [==============================] - 0s 9ms/step - loss: 3.4212e-05 -
accuracy: 1.0000
Epoch 21/50
1/1 [==============================] - 0s 11ms/step - loss: 6.1989e-06 -
accuracy: 1.0000
Epoch 22/50
1/1 [==============================] - 0s 11ms/step - loss: 1.4305e-06 -
accuracy: 1.0000
Epoch 23/50
1/1 [==============================] - 0s 9ms/step - loss: 3.5763e-07 -
accuracy: 1.0000
Epoch 24/50
1/1 [==============================] - 0s 10ms/step - loss: 1.1921e-07 -
accuracy: 1.0000
Epoch 25/50
1/1 [==============================] - 0s 9ms/step - loss: 1.1921e-07 -
accuracy: 1.0000
Epoch 26/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 27/50
1/1 [==============================] - 0s 11ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 28/50
1/1 [==============================] - 0s 112ms/step - loss: 0.0000e+00 -

4
accuracy: 1.0000
Epoch 29/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 30/50
1/1 [==============================] - 0s 25ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 31/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 32/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 33/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 34/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 35/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 36/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 37/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 38/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 39/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 40/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 41/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 42/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 43/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 44/50
1/1 [==============================] - 0s 11ms/step - loss: 0.0000e+00 -

5
 accuracy: 1.0000
Epoch 45/50
1/1 [==============================] - 0s 9ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 46/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 47/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 48/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 49/50
1/1 [==============================] - 0s 8ms/step - loss: 0.0000e+00 -
accuracy: 1.0000
Epoch 50/50
1/1 [==============================] - 0s 10ms/step - loss: 0.0000e+00 -
accuracy: 1.0000

[6]: <keras.src.callbacks.History at 0x1ed10b4dc90>

[7]: # Evaluate the model

loss, accuracy = model.evaluate(X_test, y_test)
print(f"Test Accuracy: {accuracy:.2f}")

1/1 [==============================] - 1s 638ms/step - loss: 18.4027 - accuracy:

0.0000e+00
Test Accuracy: 0.00