DL p

 INDEX

Sr.no	Practicals	Date	Sign
1.	To perform Matrix Multiplication and find Eigenvalues and Eigenvectors using TensorFlow in Google Colab.	10/07/2026
2.	To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras to solve the XOR (Exclusive OR) problem and understand why hidden layers are essential for solving non-linear classification problems.	24/07/2026
3.	To implement a Deep Neural Network (DNN) using TensorFlow/Keras for performing a Binary Classification task.	31/07/2026
4(a).	To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for performing Multiclass Classification and predicting the class of a given input.	07/08/2026
4(b).	To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for multiclass classification and predict the probability of each class.	21/08/2026
4(c).	To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for performing Linear Regression and predicting continuous values.	04/09/2026
5.	To evaluate a Deep Feed Forward Neural Network (DFFNN) for Regression and Multiclass Classification using K-Fold Cross-Validation.	04/09/2026
6.	To implement Regularization in a Deep Feed Forward Neural Network using TensorFlow/Keras to reduce overfitting in binary classification.	11/09/2026
7.	To implement a Recurrent Neural Network (RNN) using TensorFlow/Keras for sequence analysis and stock price prediction.	11/09/2026

https://chatgpt.com/s/t_6ad7729bb9308191b10cce8cd2e74a23

Practical 1

Aim :- 

To perform Matrix Multiplication and find Eigenvalues and Eigenvectors using TensorFlow in Google Colab.

Code  & output :  

import tensorflow as tf

A = tf.constant([[1.0, 2.0],

                 [3.0, 4.0]])

B = tf.constant([[5.0, 6.0],

                 [7.0, 8.0]])

print("Matrix A:")

print(A.numpy())

print("Matrix B:")

print(B.numpy())

result = tf.matmul(A, B)

print("Matrix Multiplication:")

print(result.numpy())

C = tf.constant([[2.0, 1.0],

                 [1.0, 2.0]])

eigenvalues, eigenvectors = tf.linalg.eigh(C)

print("Eigenvalues:")

print(eigenvalues.numpy())

print("Eigenvectors:")

print(eigenvectors.numpy())

Practical No. 2

Aim:- 

To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras to

solve the XOR (Exclusive OR) problem and understand why hidden layers are essential for

solving non-linear classification problems.

Code & output : 

import numpy as np import tensorflow as tf from tensorflow.keras.models import Sequential from tensorflow.keras.layers import Dense tf.keras.utils.set_random_seed(42) X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]], dtype=float) y = np.array([0, 1, 1, 0]) model = Sequential([ tf.keras.Input(shape=(2,)), Dense(8, activation='tanh'), Dense(8, activation='tanh'), Dense(1, activation='sigmoid') ]) model.compile( optimizer=tf.keras.optimizers.Adam(learning_rate=0.05), loss='binary_crossentropy', metrics=['accuracy'] ) model.fit(X, y, epochs=2000, verbose=0) predictions = model.predict(X, verbose=0) print("XOR Predictions:") for i in range(len(X)): print(X[i], "=", predictions[i][0]) print("Final Predicted Classes:") print((predictions >= 0.5).astype(int).flatten())

Practical No. 3

Aim

To implement a Deep Neural Network (DNN) using TensorFlow/Keras for performing a

Binary Classification task.

Code & output : 

import numpy as np

import tensorflow as tf

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense

X = np.array([[1], [2], [3], [4], [5],

              [6], [7], [8], [9], [10]], dtype=float)

y = np.array([0, 0, 0, 0, 0,

              1, 1, 1, 1, 1])

model = Sequential([

    tf.keras.Input(shape=(1,)),

    Dense(16, activation='relu'),

    Dense(8, activation='relu'),

    Dense(1, activation='sigmoid')

])

model.compile(

    optimizer='adam',

    loss='binary_crossentropy',

    metrics=['accuracy']

)

model.fit(X, y, epochs=500, verbose=0)

test = np.array([[3], [7], [9]], dtype=float)

predictions = model.predict(test, verbose=0)

for i in range(len(test)):

    print("Input:", test[i][0])

    print("Probability:", predictions[i][0])

    print("Predicted Class:", int(predictions[i][0] >= 0.5))

Practical No. 4(a)

Aim: - 

To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for

performing Multiclass Classification and predicting the class of a given input.

Code & output : -

import numpy as np import tensorflow as tf from tensorflow.keras.models import Sequential from tensorflow.keras.layers import Dense X = np.array([[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]], dtype=float) y = np.array([0, 0, 0, 1, 1, 1, 1, 2, 2, 2]) model = Sequential([ tf.keras.Input(shape=(1,)), Dense(16, activation='relu'), Dense(16, activation='relu'), Dense(3, activation='softmax') ]) model.compile( optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'] ) model.fit(X, y, epochs=500, verbose=0) test = np.array([[8.0]]) prediction = model.predict(test, verbose=0) classes = ['Poor', 'Average', 'Excellent'] predicted_class = np.argmax(prediction) print("Input:", test[0][0]) print("Predicted Class:", classes[predicted_class])

Practical No. 4(b)

Aim:- 

To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for

multiclass classification and predict the probability of each class.

Code & output :- 

import numpy as np

import tensorflow as tf

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense

X = np.array([[1], [2], [3], [4], [5],

              [6], [7], [8], [9], [10]], dtype=float)

y = np.array([0, 0, 0, 1, 1,

              1, 1, 2, 2, 2])

model = Sequential([

    tf.keras.Input(shape=(1,)),

    Dense(16, activation='relu'),

    Dense(16, activation='relu'),

    Dense(3, activation='softmax')

])

model.compile(

    optimizer='adam',

    loss='sparse_categorical_crossentropy',

    metrics=['accuracy']

)

model.fit(X, y, epochs=500, verbose=0)

test = np.array([[6.0]])

prediction = model.predict(test, verbose=0)[0]

classes = ['Poor', 'Average', 'Excellent']

print("Class Probabilities:")

for i in range(len(classes)):

    print(classes[i], ":", prediction[i])

predicted_class = np.argmax(prediction)

print("Predicted Class:", classes[predicted_class])

Practical No. 4(c)

Aim

To implement a Deep Feed Forward Neural Network (DFFNN) using TensorFlow/Keras for

performing Linear Regression and predicting continuous values.

Code :

import numpy as np

import tensorflow as tf

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense

X = np.array([[1], [2], [3], [4], [5],

              [6], [7], [8], [9], [10]], dtype=float)

y = np.array([2, 4, 6, 8, 10,

              12, 14, 16, 18, 20], dtype=float)

model = Sequential([

    tf.keras.Input(shape=(1,)),

    Dense(16, activation='relu'),

    Dense(8, activation='relu'),

    Dense(1, activation='linear')

])

model.compile(

    optimizer='adam',

    loss='mean_squared_error'

)

model.fit(X, y, epochs=1000, verbose=0)

test = np.array([[12.0]])

prediction = model.predict(test, verbose=0)

print("Input Value:", test[0][0])

print("Predicted Continuous Value:", prediction[0][0])

Practical No. 5

Aim: To evaluate a Deep Feed Forward Neural Network (DFFNN) for Regression and Multiclass

Classification using K-Fold Cross-Validation.

Code & output:

import numpy as np

import tensorflow as tf

from sklearn.model_selection import KFold

from sklearn.preprocessing import StandardScaler

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense

X = np.arange(1, 51, dtype=float).reshape(-1, 1)

y = 2 * X.flatten() + 5

kf = KFold(

    n_splits=5,

    shuffle=True,

    random_state=42

)

mse_scores = []

for fold, (train_index, test_index) in enumerate(kf.split(X), 1):

    scaler = StandardScaler()

    X_train = scaler.fit_transform(X[train_index])

    X_test = scaler.transform(X[test_index])

    model = Sequential([

        tf.keras.Input(shape=(1,)),

        Dense(16, activation='relu'),

        Dense(8, activation='relu'),

        Dense(1)

    ])

    model.compile(

        optimizer='adam',

        loss='mean_squared_error'

    )

    model.fit(

        X_train,

        y[train_index],

        epochs=300,

        verbose=0

    )

    mse = model.evaluate(

        X_test,

        y[test_index],

        verbose=0

    )

    mse_scores.append(mse)

    print("Fold", fold, "MSE:", mse)

print("Average MSE:", np.mean(mse_scores))

Part B 

import numpy as np

import tensorflow as tf

from sklearn.datasets import load_iris

from sklearn.model_selection import StratifiedKFold

from sklearn.preprocessing import StandardScaler

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense

iris = load_iris()

X = iris.data

y = iris.target

kf = StratifiedKFold(

    n_splits=5,

    shuffle=True,

    random_state=42

)

accuracy_scores = []

for fold, (train_index, test_index) in enumerate(kf.split(X, y), 1):

    scaler = StandardScaler()

    X_train = scaler.fit_transform(X[train_index])

    X_test = scaler.transform(X[test_index])

    model = Sequential([

        tf.keras.Input(shape=(4,)),

        Dense(16, activation='relu'),

        Dense(8, activation='relu'),

        Dense(3, activation='softmax')

    ])

    model.compile(

        optimizer='adam',

        loss='sparse_categorical_crossentropy',

        metrics=['accuracy']

    )

    model.fit(

        X_train,

        y[train_index],

        epochs=100,

        verbose=0

    )

    loss, accuracy = model.evaluate(

        X_test,

        y[test_index],

        verbose=0

    )

    accuracy_scores.append(accuracy)

    print("Fold", fold, "Accuracy:", accuracy)

print("Average Accuracy:", np.mean(accuracy_scores))

Practical No. 6

Aim : To implement Regularization in a Deep Feed Forward Neural Network using TensorFlow/Keras to reduce overfitting in binary classification.

Code :

import numpy as np

import tensorflow as tf

from sklearn.datasets import make_classification

from sklearn.model_selection import train_test_split

from sklearn.preprocessing import StandardScaler

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense, Dropout

from tensorflow.keras.regularizers import l2

X, y = make_classification(

    n_samples=500,

    n_features=10,

    n_informative=5,

    random_state=42

)

X_train, X_test, y_train, y_test = train_test_split(

    X, y,

    test_size=0.2,

    random_state=42

)

scaler = StandardScaler()

X_train = scaler.fit_transform(X_train)

X_test = scaler.transform(X_test)

model = Sequential([

    tf.keras.Input(shape=(10,)),

    Dense(64, activation='relu', kernel_regularizer=l2(0.01)),

    Dropout(0.3),

    Dense(32, activation='relu', kernel_regularizer=l2(0.01)),

    Dropout(0.3),

    Dense(1, activation='sigmoid')

])

model.compile(

    optimizer='adam',

    loss='binary_crossentropy',

    metrics=['accuracy']

)

model.fit(

    X_train,

    y_train,

    epochs=50,

    validation_split=0.2,

    verbose=0

)

loss, accuracy = model.evaluate(

    X_test,

    y_test,

    verbose=0

)

print("Test Accuracy:", accuracy)

print("Test Loss:", loss)

prediction = model.predict(X_test[:5], verbose=0)

print("Predicted Classes:")

print((prediction >= 0.5).astype(int).flatten())

Practical No. 7

Aim:

To implement a Recurrent Neural Network (RNN) using TensorFlow/Keras for sequence analysis and stock price prediction.

Code :

import numpy as np import tensorflow as tf import matplotlib.pyplot as plt from sklearn.preprocessing import MinMaxScaler from tensorflow.keras.models import Sequential from tensorflow.keras.layers import SimpleRNN, Dense np.random.seed(42) # Generate sample stock price data prices = np.linspace(100, 200, 200) + np.random.randn(200) * 3 prices = prices.reshape(-1, 1) scaler = MinMaxScaler() scaled_prices = scaler.fit_transform(prices) X = [] y = [] for i in range(10, len(scaled_prices)): X.append(scaled_prices[i-10:i]) y.append(scaled_prices[i]) X = np.array(X) y = np.array(y) split = int(len(X) * 0.8) X_train = X[:split] X_test = X[split:] y_train = y[:split] y_test = y[split:] model = Sequential([ tf.keras.Input(shape=(10, 1)), SimpleRNN(50, activation='tanh'), Dense(1) ]) model.compile( optimizer='adam', loss='mean_squared_error' ) model.fit( X_train, y_train, epochs=100, batch_size=16, verbose=0 ) predictions = model.predict(X_test, verbose=0) actual = scaler.inverse_transform(y_test) predicted = scaler.inverse_transform(predictions) print("Actual Stock Prices:") print(actual[:5].flatten()) print("Predicted Stock Prices:") print(predicted[:5].flatten()) next_input = scaled_prices[-10:].reshape(1, 10, 1) next_price = model.predict(next_input, verbose=0) next_price = scaler.inverse_transform(next_price) print("Next Predicted Stock Price:", next_price[0][0]) plt.plot(actual, label="Actual Price") plt.plot(predicted, label="Predicted Price") plt.title("RNN Stock Price Prediction") plt.xlabel("Time") plt.ylabel("Stock Price") plt.legend() plt.show()