Scribd
Upload
6 views

CNN Image Classification With Advanced Hyperparameter Tunning.ipynb+ +Colab

This document outlines a process for image classification using a Convolutional Neural Network (CNN) with the CIFAR-10 dataset, incorporating advanced hyperparameter tuning with Keras Tuner. It details steps for importing libraries, loading and preprocessing data, defining the model architecture, and performing hyperparameter optimization. Finally, it includes instructions for training the model and visualizing the training and validation accuracy and loss.

Uploaded by

bujjikanna478
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
6 views

CNN Image Classification With Advanced Hyperparameter Tunning.ipynb+ +Colab

This document outlines a process for image classification using a Convolutional Neural Network (CNN) with the CIFAR-10 dataset, incorporating advanced hyperparameter tuning with Keras Tuner. It details steps for importing libraries, loading and preprocessing data, defining the model architecture, and performing hyperparameter optimization. Finally, it includes instructions for training the model and visualizing the training and validation accuracy and loss.

Uploaded by

bujjikanna478
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

6/1/24, 10:44 AM cifar-10-adv-hyperparameter.

ipynb - Colab

Image Classification Using CNN with the CIFAR-10 Dataset using advanced Hyperparameter Tunning

pip install keras_tuner

# Step 1: Importing necessary libraries


import tensorflow as tf
from tensorflow.keras import datasets, layers, models
from keras_tuner.tuners import RandomSearch
import matplotlib.pyplot as plt

# Step 2: Loading and preprocessing the CIFAR-10 dataset


(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()
train_images, test_images = train_images / 255.0, test_images / 255.0

# Step 3: Defining the class names for CIFAR-10


class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']

# Step 4: Define a function to build the model.


def build_model(hp):
model = models.Sequential()

# Tune the number of Convolutional Layers (1, 2 or 3)


for i in range(hp.Int('conv_layers', 1, 3)):
if i == 0:
model.add(layers.Conv2D(
filters = hp.Int('filters_' + str(i), min_value=32, max_value=128, step=16),
kernel_size = (3, 3),
activation = 'relu',
input_shape = (32, 32, 3)
))
else:
model.add(layers.Conv2D(
filters = hp.Int('filters_' + str(i), min_value=32, max_value=128, step=16),
kernel_size = (3, 3),
activation = 'relu',
padding = 'same'))

model.add(layers.MaxPooling2D(pool_size=(2, 2)))

model.add(layers.Flatten())

# Tune the number of Dense Layers (1,2 or 3)


for i in range(hp.Int('dense_layers', 1, 3)):
model.add(layers.Dense(
units = hp.Int('units_' + str(i), min_value=32, max_value=128, step=16),
activation = 'relu'))

# Tune the dropout rate


model.add(layers.Dropout(rate=hp.Float('dropout_' + str(i), min_value=0.0, max_value=0.5, step=0.1)))

# The last dense layer with 10 output units (for 10 classes of CIFAR-10 dataset)
model.add(layers.Dense(10))

# Choose an optimizer and learning rate


optimizer = tf.keras.optimizers.Adam(learning_rate=hp.Choice('learning_rate', values=[1e-2, 1e-3, 1e-4]))

model.compile(optimizer=optimizer, loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=['accuracy'])

return model

# Step 5: Define the Tuner


tuner = RandomSearch(
build_model,
objective='val_accuracy',
max_trials=10,
executions_per_trial=1,
directory='my_dir',
project_name='cifar10_tunning'
)

# Step 6: Perform the Hyperparameter search


tuner.search(train_images, train_labels, epochs=5, validation_data=(test_images, test_labels))

# Step 7: Get the best Hyperparameters


best_hps = tuner.get_best_hyperparameters(num_trials=1)[0]

# Step 8: Build the model with the best Hyperparameters and train it
model = tuner.hypermodel.build(best_hps)
history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

https://colab.research.google.com/drive/1R5DUEfhiL-r-IOycZfZuoi8SYi-qHJCC#scrollTo=LTmwLdH8HtoZ&printMode=true 1/3
6/1/24, 10:44 AM cifar-10-adv-hyperparameter.ipynb - Colab

# Step 9: Plotting training & validation accuracy and loss values


plt.figure(figsize=(12, 4))

plt.subplot(1, 2, 1)
plt.plot(history.history['accuracy'], label='accuracy')
plt.plot(history.history['val_accuracy'], label = 'val_accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.ylim([0, 1])
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.grid()

plt.subplot(1, 2, 2)
plt.plot(history.history['loss'], label='loss')
plt.plot(history.history['val_loss'], label = 'val_loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.grid()

plt.show()

Trial 10 Complete [00h 01m 14s]


val_accuracy: 0.6549000144004822

Best val_accuracy So Far: 0.6758999824523926


Total elapsed time: 00h 10m 42s
Epoch 1/10
1563/1563 [==============================] - 15s 8ms/step - loss: 1.8427 - accuracy:
Epoch 2/10
1563/1563 [==============================] - 13s 8ms/step - loss: 1.3892 - accuracy:
Epoch 3/10
1563/1563 [==============================] - 13s 8ms/step - loss: 1.2268 - accuracy:
Epoch 4/10
1563/1563 [==============================] - 12s 8ms/step - loss: 1.1221 - accuracy:
Epoch 5/10
1563/1563 [==============================] - 13s 8ms/step - loss: 1.0406 - accuracy:
Epoch 6/10
1563/1563 [==============================] - 12s 8ms/step - loss: 0.9794 - accuracy:
Epoch 7/10
1563/1563 [==============================] - 12s 8ms/step - loss: 0.9192 - accuracy:
Epoch 8/10
1563/1563 [==============================] - 12s 8ms/step - loss: 0.8817 - accuracy:
Epoch 9/10
1563/1563 [==============================] - 12s 8ms/step - loss: 0.8445 - accuracy:
Epoch 10/10
1563/1563 [==============================] - 12s 8ms/step - loss: 0.8044 - accuracy:

https://colab.research.google.com/drive/1R5DUEfhiL-r-IOycZfZuoi8SYi-qHJCC#scrollTo=LTmwLdH8HtoZ&printMode=true 2/3

You might also like