AI Using Python

Hands-on Exercise No. 4

Task 1

import tensorflow as tf
from tensorflow import keras
import [Link] as plt

# -------------------------------
# Load Fashion-MNIST Dataset
# -------------------------------
(X_train, y_train), (X_test, y_test) = [Link].fashion_mnist.load_data()

print("Training Images Shape :", X_train.shape)

print("Training Labels Shape :", y_train.shape)
print("Testing Images Shape :", X_test.shape)
print("Testing Labels Shape :", y_test.shape)

# -------------------------------
# Normalize Pixel Values
# -------------------------------
X_train = X_train / 255.0
X_test = X_test / 255.0

# -------------------------------
# Reshape Images for CNN
# (28, 28) -> (28, 28, 1)
# -------------------------------
X_train = X_train.reshape(-1, 28, 28, 1)
X_test = X_test.reshape(-1, 28, 28, 1)

print("\nAfter Reshaping")
print("Training Images :", X_train.shape)
print("Testing Images :", X_test.shape)

# -------------------------------
# Fashion-MNIST Class Names
# -------------------------------
class_names = [
"T-shirt/Top",
"Trouser",
"Pullover",
"Dress",
"Coat",
"Sandal",
"Shirt",
"Sneaker",
"Bag",
"Ankle Boot"
]

# -------------------------------
# Display Sample Images
# -------------------------------
[Link](figsize=(10, 6))

for i in range(10):
[Link](2, 5, i + 1)
[Link](X_train[i].reshape(28, 28), cmap="gray")
[Link](class_names[y_train[i]])
[Link]("off")

plt.tight_layout()
[Link]()
Task 2
Task 3: Compile and Train the CNN

import tensorflow as tf
from tensorflow import keras

# -------------------------------
# Load Dataset
# -------------------------------
(X_train, y_train), (X_test, y_test) = [Link].fashion_mnist.load_data()

# Normalize
X_train = X_train / 255.0
X_test = X_test / 255.0

# Reshape for CNN

X_train = X_train.reshape(-1, 28, 28, 1)
X_test = X_test.reshape(-1, 28, 28, 1)

# -------------------------------
# Build CNN Model
# -------------------------------
model = [Link]([
[Link](shape=(28, 28, 1)),
[Link].Conv2D(32, (3,3), activation="relu"),
[Link].MaxPooling2D((2,2)),
[Link].Conv2D(64, (3,3), activation="relu"),
[Link].MaxPooling2D((2,2)),
[Link](),
[Link](128, activation="relu"),
 [Link](10, activation="softmax")
])

# -------------------------------
# Compile Model
# -------------------------------
[Link](
optimizer="adam",
loss="sparse_categorical_crossentropy",
metrics=["accuracy"]
)

# -------------------------------
# Train Model
# -------------------------------
history = [Link](
X_train,
y_train,
epochs=5,
validation_data=(X_test, y_test)
)

print("\nTraining Completed Successfully!")

Task 4: Evaluate the Model and Make Predictions.

import tensorflow as tf
from tensorflow import keras
import [Link] as plt

# -------------------------------
# Load Dataset
# -------------------------------
(X_train, y_train), (X_test, y_test) = [Link].fashion_mnist.load_data()

# Normalize
X_train = X_train / 255.0
X_test = X_test / 255.0

# Reshape for CNN

X_train = X_train.reshape(-1, 28, 28, 1)
X_test = X_test.reshape(-1, 28, 28, 1)

# -------------------------------
# Build CNN Model
# -------------------------------
model = [Link]([
[Link](shape=(28, 28, 1)),
[Link].Conv2D(32, (3, 3), activation="relu"),
[Link].MaxPooling2D((2, 2)),
[Link].Conv2D(64, (3, 3), activation="relu"),
[Link].MaxPooling2D((2, 2)),
[Link](),
[Link](128, activation="relu"),
[Link](10, activation="softmax")
])

# -------------------------------
# Compile Model
# -------------------------------
[Link](
optimizer="adam",
loss="sparse_categorical_crossentropy",
metrics=["accuracy"]
)

# -------------------------------
# Train Model
# -------------------------------
[Link](
X_train,
y_train,
epochs=5,
validation_data=(X_test, y_test),
verbose=1
)
# -------------------------------
# Evaluate Model
# -------------------------------
test_loss, test_accuracy = [Link](X_test, y_test, verbose=0)

print("\n==============================")
print(f"Test Loss : {test_loss:.4f}")
print(f"Test Accuracy : {test_accuracy * 100:.2f}%")
print("==============================")

# -------------------------------
# Fashion-MNIST Labels
# -------------------------------
class_names = [
"T-shirt/Top",
"Trouser",
"Pullover",
"Dress",
"Coat",
"Sandal",
"Shirt",
"Sneaker",
"Bag",
"Ankle Boot"
]

# -------------------------------
# Predict First 10 Images
# -------------------------------
predictions = [Link](X_test[:10])

[Link](figsize=(12, 6))

for i in range(10):
[Link](2, 5, i + 1)

[Link](X_test[i].reshape(28, 28), cmap="gray")

predicted_label = class_names[predictions[i].argmax()]
actual_label = class_names[y_test[i]]

[Link](f"P: {predicted_label}\nA: {actual_label}", fontsize=9)

[Link]("off")

plt.tight_layout()
[Link]()
Task 5

import tensorflow as tf
from tensorflow import keras
import [Link] as plt
# ==========================================================
# Task 1: Load and Prepare Fashion-MNIST Dataset
# ==========================================================

(X_train, y_train), (X_test, y_test) = [Link].fashion_mnist.load_data()

print("Training Images Shape :", X_train.shape)

print("Training Labels Shape :", y_train.shape)
print("Testing Images Shape :", X_test.shape)
print("Testing Labels Shape :", y_test.shape)

# Normalize pixel values

X_train = X_train / 255.0
X_test = X_test / 255.0

# Reshape images for CNN

X_train = X_train.reshape(-1, 28, 28, 1)
X_test = X_test.reshape(-1, 28, 28, 1)

print("\nAfter Reshaping")
print("Training Images :", X_train.shape)
print("Testing Images :", X_test.shape)

# Fashion-MNIST Classes
class_names = [
"T-shirt/Top",
"Trouser",
"Pullover",
"Dress",
"Coat",
"Sandal",
"Shirt",
"Sneaker",
"Bag",
"Ankle Boot"
]

# Display Sample Images

[Link](figsize=(10,6))

for i in range(10):
[Link](2,5,i+1)
[Link](X_train[i].reshape(28,28), cmap="gray")
[Link](class_names[y_train[i]])
[Link]("off")

plt.tight_layout()
[Link]()

# ==========================================================
# Task 2: Build CNN Model
# ==========================================================

model = [Link]([

[Link].Conv2D(
filters=32,
kernel_size=(3,3),
activation='relu',
input_shape=(28,28,1)
),

[Link].MaxPooling2D((2,2)),

[Link].Conv2D(
filters=64,
kernel_size=(3,3),
activation='relu'
),

[Link].MaxPooling2D((2,2)),

[Link](),

[Link](
128,
activation='relu'
),

[Link](
10,
activation='softmax'
)
])

print("\n================ MODEL SUMMARY ================\n")

[Link]()

# ==========================================================
# Task 3: Compile and Train Model
# ==========================================================

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

history = [Link](
X_train,
y_train,
 epochs=10,
validation_data=(X_test, y_test)
)

# ==========================================================
# Task 4: Evaluate Model
# ==========================================================

test_loss, test_accuracy = [Link](X_test, y_test)

print("\n================ RESULTS ================")

print("Test Loss :", test_loss)
print("Test Accuracy :", test_accuracy)

# ==========================================================
# Task 5: Visualization and Analysis
# ==========================================================

# Accuracy Graph
[Link](figsize=(12,5))

[Link](1,2,1)
[Link]([Link]['accuracy'], label='Training Accuracy')
[Link]([Link]['val_accuracy'], label='Validation Accuracy')
[Link]("Training vs Validation Accuracy")
[Link]("Epoch")
[Link]("Accuracy")
[Link]()

# Loss Graph
[Link](1,2,2)
[Link]([Link]['loss'], label='Training Loss')
[Link]([Link]['val_loss'], label='Validation Loss')
[Link]("Training vs Validation Loss")
[Link]("Epoch")
[Link]("Loss")
[Link]()

plt.tight_layout()
[Link]()

# ==========================================================
# Predict Test Images
# ==========================================================

predictions = [Link](X_test)

predicted_labels = [Link](axis=1)

# ==========================================================
# Display Predictions
# ==========================================================

[Link](figsize=(12,6))

for i in range(10):

[Link](2,5,i+1)

[Link](X_test[i].reshape(28,28), cmap="gray")

[Link](
f"Pred: {class_names[predicted_labels[i]]}\nTrue: {class_names[y_test[i]]}",
fontsize=8
)

[Link]("off")

plt.tight_layout()
[Link]()

print("\nTask 5 Completed Successfully!")