Scribd
Upload
8 views2 pages

Prac 12 ANN

The document outlines a TensorFlow implementation for training a Convolutional Neural Network (CNN) on the MNIST dataset. It includes steps for data loading, preprocessing, model definition, compilation, training, and evaluation, achieving a test accuracy of 99.04%. The model consists of multiple convolutional and pooling layers, followed by dense layers, and is trained over 10 epochs.

Uploaded by

pjpatel
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views2 pages

Prac 12 ANN

The document outlines a TensorFlow implementation for training a Convolutional Neural Network (CNN) on the MNIST dataset. It includes steps for data loading, preprocessing, model definition, compilation, training, and evaluation, achieving a test accuracy of 99.04%. The model consists of multiple convolutional and pooling layers, followed by dense layers, and is trained over 10 epochs.

Uploaded by

pjpatel
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

4/15/25, 11:59 AM Untitled3

In [1]: #SL II(ANN) - Assignment=12 (Group C2)

import tensorflow as tf
from tensorflow.keras.datasets import mnist
from tensorflow.keras.layers import Conv2D, Dense, Flatten, MaxPooling2D
from tensorflow.keras.models import Sequential
from tensorflow.keras.optimizers import Adam

# Load the dataset


(x_train, y_train), (x_test, y_test) = mnist.load_data()

# Preprocess the data


x_train = x_train / 255.0
x_test = x_test / 255.0

# Reshape the input data to have a channel dimension


x_train = x_train.reshape((x_train.shape[0], 28, 28, 1))
x_test = x_test.reshape((x_test.shape[0], 28, 28, 1))

# Define the CNN model


model = Sequential([
Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),
MaxPooling2D((2, 2)),
Conv2D(64, (3, 3), activation='relu'),
MaxPooling2D((2, 2)),
Flatten(),
Dense(64, activation='relu'),
Dense(10, activation='softmax')
])

# Compile the model


model.compile(
optimizer=Adam(learning_rate=0.001),
loss='sparse_categorical_crossentropy',
metrics=['accuracy']
)

# Train the model


model.fit(x_train, y_train, validation_split=0.1, epochs=10, batch_size=32)

# Evaluate the model


test_loss, test_acc = model.evaluate(x_test, y_test)
print(f"Test loss: {test_loss:.4f}, Test accuracy: {test_acc:.4f}")

localhost:8888/nbconvert/html/Untitled3.ipynb?download=false 1/2
4/15/25, 11:59 AM Untitled3
Epoch 1/10
1688/1688 [==============================] - 14s 8ms/step - loss: 0.1610 - accuracy: 0.9514 - val_
loss: 0.0625 - val_accuracy: 0.9797
Epoch 2/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0556 - accuracy: 0.9826 - val_
loss: 0.0409 - val_accuracy: 0.9885
Epoch 3/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0385 - accuracy: 0.9881 - val_
loss: 0.0400 - val_accuracy: 0.9885
Epoch 4/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0287 - accuracy: 0.9909 - val_
loss: 0.0327 - val_accuracy: 0.9902
Epoch 5/10
1688/1688 [==============================] - 14s 8ms/step - loss: 0.0225 - accuracy: 0.9930 - val_
loss: 0.0329 - val_accuracy: 0.9913
Epoch 6/10
1688/1688 [==============================] - 14s 8ms/step - loss: 0.0156 - accuracy: 0.9947 - val_
loss: 0.0328 - val_accuracy: 0.9900
Epoch 7/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0132 - accuracy: 0.9957 - val_
loss: 0.0399 - val_accuracy: 0.9902
Epoch 8/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0104 - accuracy: 0.9967 - val_
loss: 0.0418 - val_accuracy: 0.9892
Epoch 9/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0081 - accuracy: 0.9974 - val_
loss: 0.0536 - val_accuracy: 0.9903
Epoch 10/10
1688/1688 [==============================] - 13s 8ms/step - loss: 0.0086 - accuracy: 0.9969 - val_
loss: 0.0388 - val_accuracy: 0.9920
313/313 [==============================] - 1s 3ms/step - loss: 0.0424 - accuracy: 0.9904
Test loss: 0.0424, Test accuracy: 0.9904

In [ ]:

localhost:8888/nbconvert/html/Untitled3.ipynb?download=false 2/2

You might also like