Scribd
Upload
11 views2 pages

CIFAR-10 CNN Model Training Guide

The document outlines a TensorFlow script for training a Convolutional Neural Network (CNN) on the CIFAR-10 dataset, achieving a test accuracy of 70.61%. It includes steps for data loading, model definition, compilation, training, evaluation, and making predictions on new images. The model predicts the class of a new image, returning 'deer' as the predicted class index 4.

Uploaded by

ramsai
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
11 views2 pages

CIFAR-10 CNN Model Training Guide

The document outlines a TensorFlow script for training a Convolutional Neural Network (CNN) on the CIFAR-10 dataset, achieving a test accuracy of 70.61%. It includes steps for data loading, model definition, compilation, training, evaluation, and making predictions on new images. The model predicts the class of a new image, returning 'deer' as the predicted class index 4.

Uploaded by

ramsai
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

In [1]:

import tensorflow as tf
from [Link] import datasets, layers, models
import [Link] as plt
import numpy as np
from PIL import Image

# Step 1: Load and preprocess the CIFAR-10 dataset


(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()

# Normalize pixel values to be between 0 and 1


train_images, test_images = train_images / 255.0, test_images / 255.0

# Step 2: Define the CNN model


model = [Link]([
layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
[Link](),
[Link](64, activation='relu'),
[Link](10) # Output layer with 10 classes
])

# Step 3: Compile the model


[Link](optimizer='adam',
loss=[Link](from_logits=True),
metrics=['accuracy'])

# Step 4: Train the model


history = [Link](train_images, train_labels, epochs=10,
validation_data=(test_images, test_labels))

# Step 5: Evaluate the model


test_loss, test_acc = [Link](test_images, test_labels, verbose=2)
print(f"Test accuracy: {test_acc*100:.2f}%")

# Step 6: Make predictions on a new image


def predict_new_image(image_path, model):
img = [Link](image_path)
img = [Link]((32, 32))
img_array = [Link](img) / 255.0
img_array = np.expand_dims(img_array, axis=0) # Add batch dimension
predictions = [Link](img_array)
class_index = [Link](predictions[0])
return class_index

new_image_path = '[Link]'
class_index= predict_new_image(new_image_path, model)
l=["airplane","automobile","bird","cat","deer","dog","frog","horse","ship","truck"]
print(l[class_index])
print(f"Predicted class index: {class_index}")

Epoch 1/10
1563/1563 [==============================] - 48s 30ms/step - loss: 1.5246 - accuracy: 0.4450 - val_loss: 1.3093 - val_accuracy: 0.5379
Epoch 2/10
1563/1563 [==============================] - 47s 30ms/step - loss: 1.1922 - accuracy: 0.5789 - val_loss: 1.0812 - val_accuracy: 0.6116
Epoch 3/10
1563/1563 [==============================] - 45s 29ms/step - loss: 1.0344 - accuracy: 0.6366 - val_loss: 1.0275 - val_accuracy: 0.6396
Epoch 4/10
1563/1563 [==============================] - 47s 30ms/step - loss: 0.9392 - accuracy: 0.6707 - val_loss: 0.9740 - val_accuracy: 0.6582
Epoch 5/10
1563/1563 [==============================] - 48s 30ms/step - loss: 0.8719 - accuracy: 0.6952 - val_loss: 0.9012 - val_accuracy: 0.6892
Epoch 6/10
1563/1563 [==============================] - 51s 32ms/step - loss: 0.8102 - accuracy: 0.7138 - val_loss: 0.8784 - val_accuracy: 0.7001
Epoch 7/10
1563/1563 [==============================] - 48s 31ms/step - loss: 0.7664 - accuracy: 0.7299 - val_loss: 0.8704 - val_accuracy: 0.7029
Epoch 8/10
1563/1563 [==============================] - 45s 29ms/step - loss: 0.7188 - accuracy: 0.7483 - val_loss: 0.8548 - val_accuracy: 0.7056
Epoch 9/10
1563/1563 [==============================] - 46s 29ms/step - loss: 0.6822 - accuracy: 0.7596 - val_loss: 0.8431 - val_accuracy: 0.7141
Epoch 10/10
1563/1563 [==============================] - 46s 29ms/step - loss: 0.6434 - accuracy: 0.7740 - val_loss: 0.8731 - val_accuracy: 0.7061
313/313 - 3s - loss: 0.8731 - accuracy: 0.7061 - 3s/epoch - 9ms/step
Test accuracy: 70.61%
1/1 [==============================] - 0s 203ms/step
deer
Predicted class index: 4

In [ ]:

You might also like