create a neural net that convertes cartesian coordinate in polar cordinates in python

main.py
import keras
from keras.models import Sequential
from keras.layers import Dense
import numpy as np
99 chars
5 lines

main.py
# Define the number of data points
num_points = 1000

# Generate random Cartesian coordinates
x = np.random.uniform(low=-10, high=10, size=(num_points,))
y = np.random.uniform(low=-10, high=10, size=(num_points,))

# Convert Cartesian coordinates to polar coordinates
r = np.sqrt(x**2 + y**2)
theta = np.arctan2(y, x)

# Combine the input features into one array
input_data = np.vstack((x, y)).T

# Combine the output features into one array
output_data = np.vstack((r, theta)).T
480 chars
17 lines

main.py
# Initialize the neural network model
model = Sequential()

# Add the input layer with 2 neurons (for x and y coordinates)
model.add(Dense(2, input_dim=2, activation='relu'))

# Add a hidden layer with 16 neurons
model.add(Dense(16, activation='relu'))

# Add the output layer with 2 neurons (for r and theta)
model.add(Dense(2, activation=None))
347 chars
12 lines

main.py
# Compile the model
model.compile(loss='mean_squared_error', optimizer='adam')

# Train the model on the generated training data
model.fit(input_data, output_data, epochs=50, batch_size=32)
190 chars
6 lines

main.py
# Generate new test data
test_data = np.array([[2.5, 3.0], [-4.0, 7.5]])

# Use the trained model to predict polar coordinates
predicted_output = model.predict(test_data)
171 chars
6 lines