ANN PRACTICAL A1

1. import numpy as np
import matplotlib.pyplot as plt

# Define the activation functions

def sigmoid(x):
return 1 / (1 + np.exp(-x))

def relu(x):
return np.maximum(0, x)

def tanh(x):
return np.tanh(x)

# Define the x-axis range

x = np.linspace(-5, 5, 100)

# Plot the activation functions

plt.plot(x, sigmoid(x), label='Sigmoid')
plt.plot(x, relu(x), label='ReLU')
plt.plot(x, tanh(x), label='Tanh')

# Set the plot title and labels

plt.title('Activation Functions')
plt.xlabel('Input')
plt.ylabel('Output')

# Show the legend and display the plot

plt.legend()
plt.show()
 ANN PRACTICAL A4
import numpy as np
import matplotlib.pyplot as plt

# Define the perceptron function

def perceptron(x, w, b):
return np.sign(np.dot(x, w) + b)

# Define the perceptron learning algorithm

def perceptron_learning(X, Y, learning_rate, n_iterations):
n_samples, n_features = X.shape
w = np.zeros(n_features)
b = 0
for _ in range(n_iterations):
for i in range(n_samples):
x = X[i]
y = Y[i]
y_pred = perceptron(x, w, b)
if y_pred != y:
w += learning_rate * y * x
b += learning_rate * y
return w, b

# Generate some random data for binary classification

np.random.seed(0)
X = np.random.randn(200, 2)
Y = np.zeros(200)
Y[X[:, 0] + X[:, 1] > 0] = 1

# Train the perceptron on the data

w, b = perceptron_learning(X, Y, 0.1, 100)

# Plot the decision boundary

fig, ax = plt.subplots()
ax.scatter(X[:, 0], X[:, 1], c=Y, cmap=plt.cm.RdYlBu)
xlim = ax.get_xlim()
ylim = ax.get_ylim()
x = np.linspace(xlim[0], xlim[1], 100)
y = np.linspace(ylim[0], ylim[1], 100)
xx, yy = np.meshgrid(x, y)
xy = np.vstack([xx.ravel(), yy.ravel()]).T
Z = perceptron(xy, w, b).reshape(xx.shape)
ax.contour(xx, yy, Z, colors='k', levels=[-1, 0, 1], alpha=0.5)
ax.set_title('Perceptron Decision Regions')
plt.show()
 ANN PRACTICAL A5
import numpy as np

# Define the BAM function

def BAM(X, Y):
n_samples, n_features = X.shape
_, n_labels = Y.shape
W = np.zeros((n_features, n_labels))
for i in range(n_samples):
x = X[i].reshape((n_features, 1))
y = Y[i].reshape((n_labels, 1))
W += np.dot(x, y.T)
return W

# Define the bidirectional recall function

def bidirectional_recall(X, W):
n_samples, n_features = X.shape
Y = np.zeros((n_samples, W.shape[1]))
for i in range(n_samples):
x = X[i].reshape((n_features, 1))
y = np.dot(W.T, x)
y[y >= 0] = 1
y[y < 0] = -1
Y[i] = y.T
return Y

# Define the input and output patterns

X = np.array([[1, 1, -1, -1], [1, -1, 1, -1]])
Y = np.array([[1, -1], [-1, 1]])

# Train the BAM on the patterns

W = BAM(X, Y)

# Recall the output patterns from the input patterns

Y_recalled = bidirectional_recall(X, W)

# Print the output patterns

print('Input Patterns:\n', X)
print('Output Patterns:\n', Y_recalled)

OUTPUT:
Input Patterns:
[[ 1 1 -1 -1]
[ 1 -1 1 -1]]
Output Patterns:
[[ 1. -1.]
[-1. 1.]]

ANN PRACTICAL A7
import numpy as np
def sigmoid(x):
return 1 / (1 + np.exp(-x))
def sigmoid_derivative(x):
 return x * (1 - x)
class ANN:

def __init__(self, input_shape, output_shape, hidden_shape):

self.input_shape = input_shape
self.output_shape = output_shape
self.hidden_shape = hidden_shape

self.weights_ih = np.random.randn(self.input_shape, self.hidden_shape)

self.weights_ho = np.random.randn(self.hidden_shape, self.output_shape)

self.biases_h = np.zeros((1, self.hidden_shape))

self.biases_o = np.zeros((1, self.output_shape))

def forward_propagation(self, inputs):

self.hidden_layer = sigmoid(np.dot(inputs, self.weights_ih) +
self.biases_h)
self.output_layer = sigmoid(np.dot(self.hidden_layer, self.weights_ho) +
self.biases_o)

def backpropagation(self, inputs, targets, learning_rate):

error = targets - self.output_layer
d_output = error * sigmoid_derivative(self.output_layer)

error_hidden = d_output.dot(self.weights_ho.T)
d_hidden = error_hidden * sigmoid_derivative(self.hidden_layer)

self.weights_ho += self.hidden_layer.T.dot(d_output) * learning_rate

self.weights_ih += inputs.T.dot(d_hidden) * learning_rate

self.biases_o += np.sum(d_output, axis=0) * learning_rate

self.biases_h += np.sum(d_hidden, axis=0) * learning_rate

# define input and output data

X = np.array([[0, 0, 1], [0, 1, 1], [1, 0, 1], [1, 1, 1]])
y = np.array([[0], [1], [1], [0]])
# create ANN instance
ann = ANN(input_shape=3, output_shape=1, hidden_shape=4)
# train ANN
for i in range(10000):
ann.forward_propagation(X)
ann.backpropagation(X, y, 0.1)

# make predictions
ann.forward_propagation(X)
print(ann.output_layer)

OUTPUT:
[[0.0383657 ]
[0.96089964]
[0.9631848 ]
[0.03347953]]
ANN PRACTICAL B1
import numpy as np
# Define the sigmoid function
def sigmoid(x):
return 1 / (1 + np.exp(-x))
# Define the derivative of sigmoid function
def sigmoid_derivative(x):
 return x * (1 - x)

# Define the inputs and expected outputs for XOR function

X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0], [1], [1], [0]])

# Initialize the weights and biases

input_layer_size = 2
hidden_layer_size = 2
output_layer_size = 1

weights1 = np.random.randn(input_layer_size, hidden_layer_size)

weights2 = np.random.randn(hidden_layer_size, output_layer_size)

bias1 = np.random.randn(hidden_layer_size)
bias2 = np.random.randn(output_layer_size)

# Set the learning rate and number of iterations

learning_rate = 0.1
num_iterations = 100000

# Perform the training using backpropagation

for i in range(num_iterations):
# Forward propagation
hidden_layer = sigmoid(np.dot(X, weights1) + bias1)
output_layer = sigmoid(np.dot(hidden_layer, weights2) + bias2)

# Backward propagation
output_layer_error = y - output_layer
output_layer_delta = output_layer_error * sigmoid_derivative(output_layer)

hidden_layer_error = output_layer_delta.dot(weights2.T)
hidden_layer_delta = hidden_layer_error * sigmoid_derivative(hidden_layer)

# Update the weights and biases

weights2 += hidden_layer.T.dot(output_layer_delta) * learning_rate
bias2 += np.sum(output_layer_delta) * learning_rate

weights1 += X.T.dot(hidden_layer_delta) * learning_rate

bias1 += np.sum(hidden_layer_delta) * learning_rate

# Test the trained model

test_input = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
predicted_output = sigmoid(np.dot(sigmoid(np.dot(test_input, weights1) + bias1),
weights2) + bias2)

print("Predicted output:")
print(np.round(predicted_output))

OUTPUT:
Predicted output:
[[0.]
[1.]
[1.]
[0.]]

ANN PRACTICAL B2
import numpy as np

class ARTNetwork:
def __init__(self, num_input, vigilance):
self.num_input = num_input
self.vigilance = vigilance
self.weights = np.zeros((num_input,))
 def train(self, input_pattern):
while True:
net_input = np.dot(input_pattern, self.weights)
if np.sum(input_pattern) <= np.sum(net_input):
return
else:
self.weights += (input_pattern - self.vigilance * net_input) / (1 +
self.vigilance)

def predict(self, input_pattern):

net_input = np.dot(input_pattern, self.weights)
output_pattern = np.zeros((self.num_input,))
output_pattern[net_input.argmax()] = 1
return output_pattern

# create an ART network with 4 input neurons and vigilance = 0.5

art = ARTNetwork(num_input=4, vigilance=0.5)

# train the network on two input patterns

art.train(np.array([1, 0, 0, 0]))
art.train(np.array([0, 0, 1, 0]))

# predict the output for a new input pattern

output = art.predict(np.array([0, 1, 0, 0]))
print(output)

OUTPUT:
[1. 0. 0. 0.]

ANN PRACTICAL B4
import numpy as np
class HopfieldNetwork:

def __init__(self, num_neurons):

self.num_neurons = num_neurons
self.weights = np.zeros((num_neurons, num_neurons))
 def train(self, patterns):
for pattern in patterns:
self.weights += np.outer(pattern, pattern)
np.fill_diagonal(self.weights, 0)

def recall(self, pattern):

converged = False
while not converged:
activation = np.dot(self.weights, pattern)
output = np.where(activation > 0, 1, -1)
if np.array_equal(output, pattern):
converged = True
else:
pattern = output
return output
# define the patterns to be stored
patterns = [
[1, 1, -1, -1],
[-1, -1, 1, 1],
[1, -1, 1, -1],
[-1, 1, -1, 1]
]

# create a Hopfield network and train it on the patterns

network = HopfieldNetwork(4)
network.train(patterns)

# test the network by recalling the patterns

for pattern in patterns:
print("Input Pattern:", pattern)
output = network.recall(pattern)
print("Recalled Pattern:", output)
print()

OUTPUT:
Input Pattern: [1, 1, -1, -1]
Recalled Pattern: [ 1 1 -1 -1]
Input Pattern: [-1, -1, 1, 1]
Recalled Pattern: [-1 -1 1 1]
Input Pattern: [1, -1, 1, -1]
Recalled Pattern: [ 1 -1 1 -1]
Input Pattern: [-1, 1, -1, 1]
Recalled Pattern: [-1 1 -1 1]

ANN PRACTICAL B5
import torch
import torchvision
import cv2
import numpy as np

# Define the model

model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)
model.eval()
# Load an image
img = cv2.imread('test.jpg')

# Preprocess the image

img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
img = np.transpose(img, (2, 0, 1))
img = torch.from_numpy(img).float().unsqueeze(0)

# Pass the image through the model

output = model(img)[0]

# Extract the bounding boxes, scores, and class labels

boxes = output['boxes'].detach().numpy()
scores = output['scores'].detach().numpy()
labels = output['labels'].detach().numpy()

# Draw the bounding boxes on the image

for i in range(len(boxes)):
if scores[i] > 0.5:
box = boxes[i]
cv2.rectangle(img, (int(box[0]), int(box[1])), (int(box[2]), int(box[3])),
(0, 255, 0), 2)
cv2.putText(img, str(labels[i]), (int(box[0]), int(box[1]) - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)

# Display the image

img = img.squeeze().numpy()
img = np.transpose(img, (1, 2, 0))
img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
cv2.imshow('image', img)
cv2.waitKey(0)
cv2.destroyAllWindows()

OUTPUT:
 ANN PRACTICAL C1

import tensorflow as tf
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

# Load the breast cancer dataset

data = load_breast_cancer()

# Split the data into training and testing sets

X_train, X_test, y_train, y_test = train_test_split(data.data, data.target,
test_size=0.2)

# Scale the data

scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

# Define the neural network architecture

model = tf.keras.models.Sequential([
tf.keras.layers.Dense(32, activation='relu', input_shape=(X_train.shape[1],)),
tf.keras.layers.Dense(1, activation='sigmoid')
])

# Compile the model

model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

# Train the model

model.fit(X_train, y_train, epochs=50, batch_size=32, validation_data=(X_test, y_test))

# Evaluate the model on the test set

loss, accuracy = model.evaluate(X_test, y_test)
print(f'Test set loss: {loss:.4f}, accuracy: {accuracy:.4f}')

# Train a logistic regression model using TensorFlow

logistic_model = tf.keras.models.Sequential([
tf.keras.layers.Dense(1, activation='sigmoid', input_shape=(X_train.shape[1],))
])
logistic_model.compile(optimizer='adam', loss='binary_crossentropy',
metrics=['accuracy'])
logistic_model.fit(X_train, y_train, epochs=50, batch_size=32, validation_data=(X_test,
y_test))
loss, accuracy = logistic_model.evaluate(X_test, y_test)
print(f'Test set loss: {loss:.4f}, accuracy: {accuracy:.4f}')

OUTPUT:
Epoch 1/50
15/15 [==============================] - 1s 27ms/step - loss: 0.7860 - accuracy: 0.4198 - val_loss: 0.7002 -
val_accuracy: 0.4298
Epoch 2/50
15/15 [==============================] - 0s 5ms/step - loss: 0.5435 - accuracy: 0.7011 - val_loss: 0.4789 -
val_accuracy: 0.8596
Epoch 3/50
15/15 [==============================] - 0s 4ms/step - loss: 0.3941 - accuracy: 0.8593 - val_loss: 0.3441 -
val_accuracy: 0.9123
Epoch 4/50
15/15 [==============================] - 0s 4ms/step - loss: 0.3036 - accuracy: 0.9143 - val_loss: 0.2644 -
val_accuracy: 0.9298
Epoch 5/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2487 - accuracy: 0.9363 - val_loss: 0.2133 -
val_accuracy: 0.9474
Epoch 6/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2128 - accuracy: 0.9407 - val_loss: 0.1795 -
val_accuracy: 0.9561
Epoch 7/50
15/15 [==============================] - 0s 8ms/step - loss: 0.1871 - accuracy: 0.9407 - val_loss: 0.1582 -
val_accuracy: 0.9561
Epoch 8/50
15/15 [==============================] - 0s 8ms/step - loss: 0.1698 - accuracy: 0.9429 - val_loss: 0.1415 -
val_accuracy: 0.9561
Epoch 9/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1559 - accuracy: 0.9495 - val_loss: 0.1281 -
val_accuracy: 0.9561
Epoch 10/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1446 - accuracy: 0.9538 - val_loss: 0.1176 -
val_accuracy: 0.9561
Epoch 11/50
15/15 [==============================] - 0s 3ms/step - loss: 0.1353 - accuracy: 0.9560 - val_loss: 0.1094 -
val_accuracy: 0.9561
Epoch 12/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1277 - accuracy: 0.9560 - val_loss: 0.1027 -
val_accuracy: 0.9737
Epoch 13/50
15/15 [==============================] - 0s 3ms/step - loss: 0.1210 - accuracy: 0.9604 - val_loss: 0.0966 -
val_accuracy: 0.9737
Epoch 14/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1152 - accuracy: 0.9604 - val_loss: 0.0911 -
val_accuracy: 0.9825
Epoch 15/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1103 - accuracy: 0.9648 - val_loss: 0.0866 -
val_accuracy: 0.9825
Epoch 16/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1054 - accuracy: 0.9670 - val_loss: 0.0817 -
val_accuracy: 0.9825
Epoch 17/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1016 - accuracy: 0.9670 - val_loss: 0.0772 -
val_accuracy: 0.9825
Epoch 18/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0978 - accuracy: 0.9670 - val_loss: 0.0742 -
val_accuracy: 0.9825
Epoch 19/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0947 - accuracy: 0.9670 - val_loss: 0.0717 -
val_accuracy: 0.9825
Epoch 20/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0916 - accuracy: 0.9714 - val_loss: 0.0690 -
val_accuracy: 0.9825
Epoch 21/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0888 - accuracy: 0.9714 - val_loss: 0.0667 -
val_accuracy: 0.9825
Epoch 22/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0863 - accuracy: 0.9714 - val_loss: 0.0647 -
val_accuracy: 0.9825
Epoch 23/50
15/15 [==============================] - 0s 5ms/step - loss: 0.0837 - accuracy: 0.9736 - val_loss: 0.0625 -
val_accuracy: 0.9825
Epoch 24/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0817 - accuracy: 0.9758 - val_loss: 0.0612 -
val_accuracy: 0.9825
Epoch 25/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0797 - accuracy: 0.9758 - val_loss: 0.0586 -
val_accuracy: 0.9912
Epoch 26/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0776 - accuracy: 0.9780 - val_loss: 0.0578 -
val_accuracy: 0.9912
Epoch 27/50
15/15 [==============================] - 0s 3ms/step - loss: 0.0759 - accuracy: 0.9780 - val_loss: 0.0563 -
val_accuracy: 0.9912
Epoch 28/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0743 - accuracy: 0.9780 - val_loss: 0.0563 -
val_accuracy: 1.0000
Epoch 29/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0728 - accuracy: 0.9780 - val_loss: 0.0549 -
val_accuracy: 1.0000
Epoch 30/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0714 - accuracy: 0.9780 - val_loss: 0.0526 -
val_accuracy: 1.0000
Epoch 31/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0701 - accuracy: 0.9780 - val_loss: 0.0509 -
val_accuracy: 1.0000
Epoch 32/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0685 - accuracy: 0.9780 - val_loss: 0.0516 -
val_accuracy: 1.0000
Epoch 33/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0674 - accuracy: 0.9802 - val_loss: 0.0492 -
val_accuracy: 1.0000
Epoch 34/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0661 - accuracy: 0.9802 - val_loss: 0.0487 -
val_accuracy: 1.0000
Epoch 35/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0648 - accuracy: 0.9802 - val_loss: 0.0476 -
val_accuracy: 1.0000
Epoch 36/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0638 - accuracy: 0.9846 - val_loss: 0.0468 -
val_accuracy: 1.0000
Epoch 37/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0626 - accuracy: 0.9846 - val_loss: 0.0450 -
val_accuracy: 1.0000
Epoch 38/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0616 - accuracy: 0.9846 - val_loss: 0.0452 -
val_accuracy: 1.0000
Epoch 39/50
15/15 [==============================] - 0s 7ms/step - loss: 0.0605 - accuracy: 0.9846 - val_loss: 0.0444 -
val_accuracy: 1.0000
Epoch 40/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0595 - accuracy: 0.9846 - val_loss: 0.0441 -
val_accuracy: 1.0000
Epoch 41/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0587 - accuracy: 0.9846 - val_loss: 0.0434 -
val_accuracy: 1.0000
Epoch 42/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0578 - accuracy: 0.9846 - val_loss: 0.0449 -
val_accuracy: 1.0000
Epoch 43/50
15/15 [==============================] - 0s 7ms/step - loss: 0.0566 - accuracy: 0.9846 - val_loss: 0.0467 -
val_accuracy: 0.9825
Epoch 44/50
15/15 [==============================] - 0s 6ms/step - loss: 0.0559 - accuracy: 0.9846 - val_loss: 0.0471 -
val_accuracy: 0.9737
Epoch 45/50
15/15 [==============================] - 0s 5ms/step - loss: 0.0550 - accuracy: 0.9868 - val_loss: 0.0471 -
val_accuracy: 0.9825
Epoch 46/50
15/15 [==============================] - 0s 5ms/step - loss: 0.0542 - accuracy: 0.9868 - val_loss: 0.0457 -
val_accuracy: 0.9912
Epoch 47/50
15/15 [==============================] - 0s 3ms/step - loss: 0.0533 - accuracy: 0.9846 - val_loss: 0.0443 -
val_accuracy: 0.9912
Epoch 48/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0524 - accuracy: 0.9846 - val_loss: 0.0433 -
val_accuracy: 0.9912
Epoch 49/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0516 - accuracy: 0.9868 - val_loss: 0.0420 -
val_accuracy: 0.9912
Epoch 50/50
15/15 [==============================] - 0s 4ms/step - loss: 0.0509 - accuracy: 0.9868 - val_loss: 0.0422 -
val_accuracy: 1.0000
4/4 [==============================] - 0s 2ms/step - loss: 0.0422 - accuracy: 1.0000
Test set loss: 0.0422, accuracy: 1.0000
Epoch 1/50
15/15 [==============================] - 1s 13ms/step - loss: 1.7898 - accuracy: 0.1758 - val_loss: 1.5630 -
val_accuracy: 0.1754
Epoch 2/50
15/15 [==============================] - 0s 4ms/step - loss: 1.6253 - accuracy: 0.2022 - val_loss: 1.4085 -
val_accuracy: 0.2018
Epoch 3/50
15/15 [==============================] - 0s 4ms/step - loss: 1.4732 - accuracy: 0.2352 - val_loss: 1.2629 -
val_accuracy: 0.2895
Epoch 4/50
15/15 [==============================] - 0s 4ms/step - loss: 1.3293 - accuracy: 0.2725 - val_loss: 1.1318 -
val_accuracy: 0.3596
Epoch 5/50
15/15 [==============================] - 0s 4ms/step - loss: 1.1994 - accuracy: 0.3231 - val_loss: 1.0122 -
val_accuracy: 0.4386
Epoch 6/50
15/15 [==============================] - 0s 4ms/step - loss: 1.0812 - accuracy: 0.3736 - val_loss: 0.9042 -
val_accuracy: 0.5000
Epoch 7/50
15/15 [==============================] - 0s 4ms/step - loss: 0.9743 - accuracy: 0.4374 - val_loss: 0.8107 -
val_accuracy: 0.5614
Epoch 8/50
15/15 [==============================] - 0s 4ms/step - loss: 0.8813 - accuracy: 0.5033 - val_loss: 0.7301 -
val_accuracy: 0.6228
Epoch 9/50
15/15 [==============================] - 0s 3ms/step - loss: 0.7998 - accuracy: 0.5626 - val_loss: 0.6616 -
val_accuracy: 0.6667
Epoch 10/50
15/15 [==============================] - 0s 4ms/step - loss: 0.7293 - accuracy: 0.6198 - val_loss: 0.6016 -
val_accuracy: 0.7105
Epoch 11/50
15/15 [==============================] - 0s 3ms/step - loss: 0.6692 - accuracy: 0.6725 - val_loss: 0.5475 -
val_accuracy: 0.7807
Epoch 12/50
15/15 [==============================] - 0s 4ms/step - loss: 0.6141 - accuracy: 0.6945 - val_loss: 0.5046 -
val_accuracy: 0.8333
Epoch 13/50
15/15 [==============================] - 0s 3ms/step - loss: 0.5683 - accuracy: 0.7297 - val_loss: 0.4670 -
val_accuracy: 0.8596
Epoch 14/50
15/15 [==============================] - 0s 4ms/step - loss: 0.5289 - accuracy: 0.7626 - val_loss: 0.4348 -
val_accuracy: 0.8684
Epoch 15/50
15/15 [==============================] - 0s 4ms/step - loss: 0.4935 - accuracy: 0.7802 - val_loss: 0.4071 -
val_accuracy: 0.8684
Epoch 16/50
15/15 [==============================] - 0s 4ms/step - loss: 0.4629 - accuracy: 0.8000 - val_loss: 0.3826 -
val_accuracy: 0.8860
Epoch 17/50
15/15 [==============================] - 0s 6ms/step - loss: 0.4367 - accuracy: 0.8198 - val_loss: 0.3615 -
val_accuracy: 0.8947
Epoch 18/50
15/15 [==============================] - 0s 7ms/step - loss: 0.4132 - accuracy: 0.8484 - val_loss: 0.3430 -
val_accuracy: 0.9035
Epoch 19/50
15/15 [==============================] - 0s 6ms/step - loss: 0.3926 - accuracy: 0.8527 - val_loss: 0.3260 -
val_accuracy: 0.9035
Epoch 20/50
15/15 [==============================] - 0s 5ms/step - loss: 0.3743 - accuracy: 0.8593 - val_loss: 0.3105 -
val_accuracy: 0.9211
Epoch 21/50
15/15 [==============================] - 0s 4ms/step - loss: 0.3573 - accuracy: 0.8725 - val_loss: 0.2970 -
val_accuracy: 0.9211
Epoch 22/50
15/15 [==============================] - 0s 5ms/step - loss: 0.3424 - accuracy: 0.8747 - val_loss: 0.2844 -
val_accuracy: 0.9386
Epoch 23/50
15/15 [==============================] - 0s 5ms/step - loss: 0.3288 - accuracy: 0.8791 - val_loss: 0.2734 -
val_accuracy: 0.9386
Epoch 24/50
15/15 [==============================] - 0s 5ms/step - loss: 0.3167 - accuracy: 0.8835 - val_loss: 0.2629 -
val_accuracy: 0.9386
Epoch 25/50
15/15 [==============================] - 0s 7ms/step - loss: 0.3049 - accuracy: 0.8879 - val_loss: 0.2537 -
val_accuracy: 0.9386
Epoch 26/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2950 - accuracy: 0.8901 - val_loss: 0.2448 -
val_accuracy: 0.9386
Epoch 27/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2847 - accuracy: 0.9011 - val_loss: 0.2364 -
val_accuracy: 0.9386
Epoch 28/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2751 - accuracy: 0.9077 - val_loss: 0.2285 -
val_accuracy: 0.9386
Epoch 29/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2667 - accuracy: 0.9143 - val_loss: 0.2209 -
val_accuracy: 0.9474
Epoch 30/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2585 - accuracy: 0.9187 - val_loss: 0.2146 -
val_accuracy: 0.9474
Epoch 31/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2511 - accuracy: 0.9209 - val_loss: 0.2085 -
val_accuracy: 0.9474
Epoch 32/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2446 - accuracy: 0.9231 - val_loss: 0.2025 -
val_accuracy: 0.9561
Epoch 33/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2382 - accuracy: 0.9297 - val_loss: 0.1973 -
val_accuracy: 0.9561
Epoch 34/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2325 - accuracy: 0.9385 - val_loss: 0.1920 -
val_accuracy: 0.9561
Epoch 35/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2269 - accuracy: 0.9385 - val_loss: 0.1871 -
val_accuracy: 0.9649
Epoch 36/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2217 - accuracy: 0.9407 - val_loss: 0.1825 -
val_accuracy: 0.9649
Epoch 37/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2170 - accuracy: 0.9407 - val_loss: 0.1780 -
val_accuracy: 0.9649
Epoch 38/50
15/15 [==============================] - 0s 4ms/step - loss: 0.2125 - accuracy: 0.9429 - val_loss: 0.1739 -
val_accuracy: 0.9649
Epoch 39/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2082 - accuracy: 0.9495 - val_loss: 0.1703 -
val_accuracy: 0.9737
Epoch 40/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2041 - accuracy: 0.9516 - val_loss: 0.1665 -
val_accuracy: 0.9825
Epoch 41/50
15/15 [==============================] - 0s 5ms/step - loss: 0.2003 - accuracy: 0.9538 - val_loss: 0.1629 -
val_accuracy: 0.9825
Epoch 42/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1966 - accuracy: 0.9538 - val_loss: 0.1597 -
val_accuracy: 0.9825
Epoch 43/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1933 - accuracy: 0.9538 - val_loss: 0.1565 -
val_accuracy: 0.9825
Epoch 44/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1899 - accuracy: 0.9560 - val_loss: 0.1536 -
val_accuracy: 0.9825
Epoch 45/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1869 - accuracy: 0.9560 - val_loss: 0.1506 -
val_accuracy: 0.9825
Epoch 46/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1838 - accuracy: 0.9604 - val_loss: 0.1478 -
val_accuracy: 0.9825
Epoch 47/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1810 - accuracy: 0.9604 - val_loss: 0.1451 -
val_accuracy: 0.9825
Epoch 48/50
15/15 [==============================] - 0s 5ms/step - loss: 0.1782 - accuracy: 0.9604 - val_loss: 0.1425 -
val_accuracy: 0.9825
Epoch 49/50
15/15 [==============================] - 0s 4ms/step - loss: 0.1756 - accuracy: 0.9626 - val_loss: 0.1400 -
val_accuracy: 0.9825
Epoch 50/50
15/15 [==============================] - 0s 3ms/step - loss: 0.1731 - accuracy: 0.9626 - val_loss: 0.1378 -
val_accuracy: 0.9825
4/4 [==============================] - 0s 2ms/step - loss: 0.1378 - accuracy: 0.9825
Test set loss: 0.1378, accuracy: 0.9825

ANN PRACTICAL C2
import tensorflow as tf

# Define the model architecture

model = tf.keras.models.Sequential([
tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(28, 28, 1)),
tf.keras.layers.MaxPooling2D(2, 2),
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(128, activation='relu'),
 tf.keras.layers.Dense(10, activation='softmax')
])

# Compile the model with an optimizer, loss function, and evaluation metric
model.compile(optimizer='adam', loss='categorical_crossentropy',
metrics=['accuracy'])

# Load the MNIST dataset and preprocess the images

mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)
x_train = x_train / 255.0
x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)
x_test = x_test / 255.0
y_train = tf.keras.utils.to_categorical(y_train, 10)
y_test = tf.keras.utils.to_categorical(y_test, 10)

# Train the model

model.fit(x_train, y_train, epochs=5)

# Evaluate the model on the test set

test_loss, test_accuracy = model.evaluate(x_test, y_test)
print("Test Loss:", test_loss)
print("Test Accuracy:", test_accuracy)

OUTPUT:
Downloading data from
https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz
11490434/11490434 [==============================] - 6s 1us/step
Epoch 1/5
1875/1875 [==============================] - 21s 11ms/step - loss: 0.1547 -
accuracy: 0.9534
Epoch 2/5
1875/1875 [==============================] - 20s 11ms/step - loss: 0.0524 -
accuracy: 0.9841
Epoch 3/5
1875/1875 [==============================] - 20s 11ms/step - loss: 0.0330 -
accuracy: 0.9894
Epoch 4/5
1875/1875 [==============================] - 20s 11ms/step - loss: 0.0216 -
accuracy: 0.9934
Epoch 5/5
1875/1875 [==============================] - 21s 11ms/step - loss: 0.0146 -
accuracy: 0.9956
313/313 [==============================] - 1s 3ms/step - loss: 0.0493 - accuracy:
0.9859
Test Loss: 0.04932432249188423
Test Accuracy: 0.9858999848365784

ANN PRACTICAL C4

In [ ]: import tensorflow from tensorflow

import keras from tensorflow.keras
import Sequential from tensorflow.keras.layers
import Dense,Flatten
In [ ]: (X_train,y_train),(X_test,y_test) = keras.datasets.mnist.load_data()
In [ ]: X_test.shape
Out[ ]: (10000, 28, 28)
In [ ]: y_train
Out[ ]: array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)
In [ ]: import matplotlib.pyplot as plt plt.imshow(X_train[2])
Out[ ]: <matplotlib.image.AxesImage at 0x7f573ed9f790>

In [ ]:
X_train[0]
X_train = X_train/255
X_test = X_test/255

Out[ ]: array([[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.01176471, 0.07058824, 0.07058824,
0.07058824, 0.49411765, 0.53333333, 0.68627451, 0.10196078,
0.65098039, 1. , 0.96862745, 0.49803922, 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.11764706, 0.14117647, 0.36862745, 0.60392157, 0.66666667, 0.99215686,
0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.88235294, 0.6745098 ,
0.99215686, 0.94901961, 0.76470588, 0.25098039, 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.19215686, 0.93333333, 0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.99215686, 0.99215686, 0.99215686, 0.98431373, 0.36470588, 0.32156863,
0.32156863, 0.21960784, 0.15294118, 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
 0. , 0. , 0.07058824, 0.85882353, 0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.77647059, 0.71372549, 0.96862745, 0.94509804, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.31372549, 0.61176471,
0.41960784, 0.99215686, 0.99215686, 0.80392157, 0.04313725, 0. , 0.16862745, 0.60392157, 0. ,
0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0.05490196, 0.00392157, 0.60392157, 0.99215686, 0.35294118, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0.54509804, 0.99215686, 0.74509804, 0.00784314, 0.
, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0.04313725, 0.74509804, 0.99215686, 0.2745098 , 0.
, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.1372549 , 0.94509804, 0.88235294,
0.62745098, 0.42352941, 0.00392157, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.31764706, 0.94117647, 0.99215686,
0.99215686, 0.46666667, 0.09803922, 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.17647059, 0.72941176,
0.99215686, 0.99215686, 0.58823529, 0.10588235,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.0627451 ,
0.36470588, 0.98823529, 0.99215686, 0.73333333,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ,
0.97647059, 0.99215686, 0.97647059,
0.25098039, 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.18039216,
0.50980392, 0.71764706, 0.99215686, 0.99215686, 0.81176471,
0.00784314, 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.15294118, 0.58039216, 0.89803922,
0.99215686, 0.99215686, 0.99215686, 0.98039216, 0.71372549,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0.09411765, 0.44705882, 0.86666667, 0.99215686, 0.99215686,
0.99215686, 0.99215686, 0.78823529, 0.30588235, 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.09019608, 0.25882353, 0.83529412, 0.99215686, 0.99215686, 0.99215686,
0.99215686, 0.77647059, 0.31764706, 0.00784314, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0.07058824, 0.67058824, 0.85882353, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.76470588, 0.31372549, 0.03529412, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
, 0. , 0. , 0. , 0.21568627,
0.6745098 , 0.88627451, 0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.95686275, 0.52156863,
0.04313725, 0. , 0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0.53333333,
0.99215686, 0.99215686, 0.99215686, 0.83137255, 0.52941176, 0.51764706, 0.0627451 , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ],
[0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.
, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. ]])
In [ ]: model = Sequential()
model.add(Flatten(input_shape=(28,28))) model.add(Dense(128,activation='relu'))
model.add(Dense(32,activation='relu')) model.add(Dense(10,activation='softmax'))
In [ ]: model.summary()
Model: "sequential_5"
_________________________________________________________________
Layer (type) Output Shape Param #
================================================================= flatten_4 (Flatten)
(None, 784) 0 dense_11 (Dense) (None, 128)
100480 dense_12 (Dense) (None, 32) 4128
dense_13 (Dense) (None, 10) 330
=================================================================
Total params: 104,938
Trainable params: 104,938
Non-trainable params: 0
_________________________________________________________________
In [ ]: model.compile(loss='sparse_categorical_crossentropy',optimizer='Adam',metrics=['acc
In [ ]: history = model.fit(X_train,y_train,epochs=25,validation_split=0.2)
Epoch 1/25
1500/1500 [==============================] - 5s 3ms/step - loss: 0.2845 - accuracy:
0.9168 - val_loss: 0.1425 - val_accuracy: 0.9570
Epoch 2/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.1164 - accuracy:
0.9638 - val_loss: 0.1258 - val_accuracy: 0.9605
Epoch 3/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0800 - accuracy:
0.9754 - val_loss: 0.0912 - val_accuracy: 0.9728
Epoch 4/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0594 - accuracy:
0.9817 - val_loss: 0.1014 - val_accuracy: 0.9697
Epoch 5/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0462 - accuracy:
0.9851 - val_loss: 0.1024 - val_accuracy: 0.9687
Epoch 6/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0377 - accuracy:
0.9881 - val_loss: 0.1060 - val_accuracy: 0.9712
Epoch 7/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0313 - accuracy:
0.9898 - val_loss: 0.0960 - val_accuracy: 0.9758
Epoch 8/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0283 - accuracy:
0.9902 - val_loss: 0.1061 - val_accuracy: 0.9732
Epoch 9/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0230 - accuracy:
0.9923 - val_loss: 0.1160 - val_accuracy: 0.9713
Epoch 10/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0179 - accuracy:
0.9939 - val_loss: 0.1134 - val_accuracy: 0.9761
Epoch 11/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0177 - accuracy:
0.9937 - val_loss: 0.1204 - val_accuracy: 0.9742
Epoch 12/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0156 - accuracy:
0.9947 - val_loss: 0.1103 - val_accuracy: 0.9761
Epoch 13/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0156 - accuracy:
0.9949 - val_loss: 0.1221 - val_accuracy: 0.9743
Epoch 14/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0121 - accuracy:
0.9960 - val_loss: 0.1246 - val_accuracy: 0.9764
Epoch 15/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0122 - accuracy:
0.9958 - val_loss: 0.1362 - val_accuracy: 0.9753
Epoch 16/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0100 - accuracy:
0.9964 - val_loss: 0.1340 - val_accuracy: 0.9772
Epoch 17/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0131 - accuracy:
0.9955 - val_loss: 0.1371 - val_accuracy: 0.9752
Epoch 18/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0101 - accuracy:
0.9968 - val_loss: 0.1318 - val_accuracy: 0.9760
Epoch 19/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0116 - accuracy:
0.9962 - val_loss: 0.1577 - val_accuracy: 0.9732
Epoch 20/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0093 - accuracy:
0.9969 - val_loss: 0.1570 - val_accuracy: 0.9718
Epoch 21/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0095 - accuracy:
0.9970 - val_loss: 0.1347 - val_accuracy: 0.9772
Epoch 22/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0076 - accuracy:
0.9975 - val_loss: 0.1591 - val_accuracy: 0.9762
Epoch 23/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0079 - accuracy:
0.9975 - val_loss: 0.1919 - val_accuracy: 0.9682
Epoch 24/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0078 - accuracy:
0.9974 - val_loss: 0.1692 - val_accuracy: 0.9737
Epoch 25/25
1500/1500 [==============================] - 4s 3ms/step - loss: 0.0084 - accuracy:
0.9971 - val_loss: 0.1647 - val_accuracy: 0.9753
In [ ]: y_prob = model.predict(X_test)
In [ ]: y_pred = y_prob.argmax(axis=1)
In [ ]: from sklearn.metrics import accuracy_score accuracy_score(y_test,y_pred)
Out[ ]: 0.9755
In [ ]: plt.plot(history.history['loss']) plt.plot(history.history['val_loss'])
Out[ ]: [<matplotlib.lines.Line2D at 0x7f5737a6c590>]
 plt.plot(history.history['accuracy']) plt.plot(history.history['val_accuracy'])

In [ ]:
Out[ ]: [<matplotlib.lines.Line2D at 0x7f5737831e50>]

In [ ]: plt.imshow(X_test[1])
Out[ ]: <matplotlib.image.AxesImage at 0x7f5736844c50>
In [ ]: model.predict(X_test[1].reshape(1,28,28)).argmax(axis=1)
Out[ ]: array([2])
In [ ]: