CoCalc -- LSTM (Long Short-Term Memory) network using TensorFlow and Keras.ipynb

GitHub Repository: suyashi29/python-su
Path: blob/master/Generative AI for Intelligent Data Handling/ LSTM (Long Short-Term Memory) network using TensorFlow and Keras.ipynb
³⁰⁷⁴ views

Kernel: Python 3 (ipykernel)

LSTM (Long Short-Term Memory) is a type of recurrent neural network (RNN) designed to address the problem of capturing long-term dependencies in sequential data.

It consists of a memory cell that can maintain information over long sequences, controlled by three gates: forget gate, input gate, and output gate.
The forget gate decides what information to discard from the cell state.
The input gate decides what new information to store in the cell state.
The output gate decides what information to output from the cell state.
LSTM's ability to retain and forget information over long periods makes it effective for tasks involving sequential data with long-term dependencies.

In [1]:

import numpy as np
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense

# Generate some random data for demonstration
# Input sequence: [0.1, 0.2, 0.3, 0.4, 0.5]
# Output sequence: [0.6, 0.7, 0.8, 0.9, 1.0]

# Define input sequence
X = np.array([[[0.1], [0.2], [0.3], [0.4], [0.5]]])

# Define output sequence
y = np.array([[0.6, 0.7, 0.8, 0.9, 1.0]])

# Define and build the LSTM model
model = Sequential([
    LSTM(50, input_shape=(5, 1)),  # 50 units in LSTM layer
    Dense(5)  # Output layer
])

# Compile the model
model.compile(optimizer='adam', loss='mse')

# Print model summary
model.summary()

# Train the model
model.fit(X, y, epochs=100, verbose=1)

# Make predictions
predictions = model.predict(X)

# Print predictions
print("Predictions:")
print(predictions)

Out[1]:

Model: "sequential"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 lstm (LSTM)                 (None, 50)                10400     
                                                                 
 dense (Dense)               (None, 5)                 255       
                                                                 
=================================================================
Total params: 10,655
Trainable params: 10,655
Non-trainable params: 0
_________________________________________________________________
Epoch 1/100
1/1 [==============================] - 3s 3s/step - loss: 0.6420
Epoch 2/100
1/1 [==============================] - 0s 14ms/step - loss: 0.6311
Epoch 3/100
1/1 [==============================] - 0s 20ms/step - loss: 0.6201
Epoch 4/100
1/1 [==============================] - 0s 28ms/step - loss: 0.6091
Epoch 5/100
1/1 [==============================] - 0s 20ms/step - loss: 0.5982
Epoch 6/100
1/1 [==============================] - 0s 18ms/step - loss: 0.5872
Epoch 7/100
1/1 [==============================] - 0s 18ms/step - loss: 0.5761
Epoch 8/100
1/1 [==============================] - 0s 19ms/step - loss: 0.5650
Epoch 9/100
1/1 [==============================] - 0s 21ms/step - loss: 0.5539
Epoch 10/100
1/1 [==============================] - 0s 18ms/step - loss: 0.5426
Epoch 11/100
1/1 [==============================] - 0s 21ms/step - loss: 0.5312
Epoch 12/100
1/1 [==============================] - 0s 18ms/step - loss: 0.5196
Epoch 13/100
1/1 [==============================] - 0s 20ms/step - loss: 0.5079
Epoch 14/100
1/1 [==============================] - 0s 20ms/step - loss: 0.4959
Epoch 15/100
1/1 [==============================] - 0s 22ms/step - loss: 0.4837
Epoch 16/100
1/1 [==============================] - 0s 22ms/step - loss: 0.4711
Epoch 17/100
1/1 [==============================] - 0s 19ms/step - loss: 0.4582
Epoch 18/100
1/1 [==============================] - 0s 21ms/step - loss: 0.4449
Epoch 19/100
1/1 [==============================] - 0s 23ms/step - loss: 0.4313
Epoch 20/100
1/1 [==============================] - 0s 19ms/step - loss: 0.4171
Epoch 21/100
1/1 [==============================] - 0s 20ms/step - loss: 0.4025
Epoch 22/100
1/1 [==============================] - 0s 19ms/step - loss: 0.3874
Epoch 23/100
1/1 [==============================] - 0s 19ms/step - loss: 0.3718
Epoch 24/100
1/1 [==============================] - 0s 21ms/step - loss: 0.3555
Epoch 25/100
1/1 [==============================] - 0s 22ms/step - loss: 0.3387
Epoch 26/100
1/1 [==============================] - 0s 16ms/step - loss: 0.3214
Epoch 27/100
1/1 [==============================] - 0s 16ms/step - loss: 0.3034
Epoch 28/100
1/1 [==============================] - 0s 20ms/step - loss: 0.2848
Epoch 29/100
1/1 [==============================] - 0s 20ms/step - loss: 0.2657
Epoch 30/100
1/1 [==============================] - 0s 21ms/step - loss: 0.2461
Epoch 31/100
1/1 [==============================] - 0s 18ms/step - loss: 0.2260
Epoch 32/100
1/1 [==============================] - 0s 18ms/step - loss: 0.2056
Epoch 33/100
1/1 [==============================] - 0s 16ms/step - loss: 0.1850
Epoch 34/100
1/1 [==============================] - 0s 19ms/step - loss: 0.1643
Epoch 35/100
1/1 [==============================] - 0s 23ms/step - loss: 0.1439
Epoch 36/100
1/1 [==============================] - 0s 19ms/step - loss: 0.1239
Epoch 37/100
1/1 [==============================] - 0s 12ms/step - loss: 0.1048
Epoch 38/100
1/1 [==============================] - 0s 21ms/step - loss: 0.0868
Epoch 39/100
1/1 [==============================] - 0s 22ms/step - loss: 0.0705
Epoch 40/100
1/1 [==============================] - 0s 17ms/step - loss: 0.0562
Epoch 41/100
1/1 [==============================] - 0s 17ms/step - loss: 0.0444
Epoch 42/100
1/1 [==============================] - 0s 14ms/step - loss: 0.0353
Epoch 43/100
1/1 [==============================] - 0s 10ms/step - loss: 0.0290
Epoch 44/100
1/1 [==============================] - 0s 16ms/step - loss: 0.0255
Epoch 45/100
1/1 [==============================] - 0s 19ms/step - loss: 0.0241
Epoch 46/100
1/1 [==============================] - 0s 12ms/step - loss: 0.0243
Epoch 47/100
1/1 [==============================] - 0s 9ms/step - loss: 0.0249
Epoch 48/100
1/1 [==============================] - 0s 16ms/step - loss: 0.0253
Epoch 49/100
1/1 [==============================] - 0s 15ms/step - loss: 0.0248
Epoch 50/100
1/1 [==============================] - 0s 10ms/step - loss: 0.0232
Epoch 51/100
1/1 [==============================] - 0s 14ms/step - loss: 0.0204
Epoch 52/100
1/1 [==============================] - 0s 9ms/step - loss: 0.0168
Epoch 53/100
1/1 [==============================] - 0s 14ms/step - loss: 0.0129
Epoch 54/100
1/1 [==============================] - 0s 18ms/step - loss: 0.0091
Epoch 55/100
1/1 [==============================] - 0s 21ms/step - loss: 0.0058
Epoch 56/100
1/1 [==============================] - 0s 20ms/step - loss: 0.0032
Epoch 57/100
1/1 [==============================] - 0s 19ms/step - loss: 0.0015
Epoch 58/100
1/1 [==============================] - 0s 17ms/step - loss: 5.7073e-04
Epoch 59/100
1/1 [==============================] - 0s 21ms/step - loss: 3.4960e-04
Epoch 60/100
1/1 [==============================] - 0s 16ms/step - loss: 6.5761e-04
Epoch 61/100
1/1 [==============================] - 0s 15ms/step - loss: 0.0013
Epoch 62/100
1/1 [==============================] - 0s 18ms/step - loss: 0.0021
Epoch 63/100
1/1 [==============================] - 0s 18ms/step - loss: 0.0030
Epoch 64/100
1/1 [==============================] - 0s 39ms/step - loss: 0.0037
Epoch 65/100
1/1 [==============================] - 0s 27ms/step - loss: 0.0042
Epoch 66/100
1/1 [==============================] - 0s 21ms/step - loss: 0.0046
Epoch 67/100
1/1 [==============================] - 0s 19ms/step - loss: 0.0047
Epoch 68/100
1/1 [==============================] - 0s 13ms/step - loss: 0.0045
Epoch 69/100
1/1 [==============================] - 0s 19ms/step - loss: 0.0042
Epoch 70/100
1/1 [==============================] - 0s 32ms/step - loss: 0.0038
Epoch 71/100
1/1 [==============================] - 0s 37ms/step - loss: 0.0033
Epoch 72/100
1/1 [==============================] - 0s 16ms/step - loss: 0.0027
Epoch 73/100
1/1 [==============================] - 0s 15ms/step - loss: 0.0022
Epoch 74/100
1/1 [==============================] - 0s 15ms/step - loss: 0.0017
Epoch 75/100
1/1 [==============================] - 0s 44ms/step - loss: 0.0014
Epoch 76/100
1/1 [==============================] - 0s 14ms/step - loss: 0.0011
Epoch 77/100
1/1 [==============================] - 0s 14ms/step - loss: 8.4289e-04
Epoch 78/100
1/1 [==============================] - 0s 16ms/step - loss: 7.1271e-04
Epoch 79/100
1/1 [==============================] - 0s 19ms/step - loss: 6.4226e-04
Epoch 80/100
1/1 [==============================] - 0s 17ms/step - loss: 6.0877e-04
Epoch 81/100
1/1 [==============================] - 0s 19ms/step - loss: 5.9024e-04
Epoch 82/100
1/1 [==============================] - 0s 12ms/step - loss: 5.6884e-04
Epoch 83/100
1/1 [==============================] - 0s 13ms/step - loss: 5.3316e-04
Epoch 84/100
1/1 [==============================] - 0s 16ms/step - loss: 4.7896e-04
Epoch 85/100
1/1 [==============================] - 0s 19ms/step - loss: 4.0839e-04
Epoch 86/100
1/1 [==============================] - 0s 11ms/step - loss: 3.2830e-04
Epoch 87/100
1/1 [==============================] - 0s 15ms/step - loss: 2.4785e-04
Epoch 88/100
1/1 [==============================] - 0s 16ms/step - loss: 1.7621e-04
Epoch 89/100
1/1 [==============================] - 0s 18ms/step - loss: 1.2069e-04
Epoch 90/100
1/1 [==============================] - 0s 16ms/step - loss: 8.5577e-05
Epoch 91/100
1/1 [==============================] - 0s 15ms/step - loss: 7.1696e-05
Epoch 92/100
1/1 [==============================] - 0s 15ms/step - loss: 7.6747e-05
Epoch 93/100
1/1 [==============================] - 0s 17ms/step - loss: 9.6050e-05
Epoch 94/100
1/1 [==============================] - 0s 13ms/step - loss: 1.2359e-04
Epoch 95/100
1/1 [==============================] - 0s 14ms/step - loss: 1.5309e-04
Epoch 96/100
1/1 [==============================] - 0s 9ms/step - loss: 1.7895e-04
Epoch 97/100
1/1 [==============================] - 0s 17ms/step - loss: 1.9691e-04
Epoch 98/100
1/1 [==============================] - 0s 16ms/step - loss: 2.0448e-04
Epoch 99/100
1/1 [==============================] - 0s 16ms/step - loss: 2.0096e-04
Epoch 100/100
1/1 [==============================] - 0s 14ms/step - loss: 1.8731e-04
1/1 [==============================] - 1s 1s/step
Predictions:
[[0.60965604 0.6877619  0.77881706 0.8914886  0.99196804]]

In [ ]:

LSTM (Long Short-Term Memory) is a type of recurrent neural network (RNN) designed to address the problem of capturing long-term dependencies in sequential data.

Product

Resources

Company