NetMon is a Python tool designed for monitoring and visualizing the training process and performance of neural networks.
NetMon is a Python-based tool designed to assist data scientists and machine learning engineers in monitoring and visualizing the training process and performance of neural networks. It provides a convenient way to track training metrics and analyze data distributions during model development.
To get started with NetMon, follow the instructions below.
Before using NetMon, ensure you have the following dependencies installed:
matplotlibnumpypandas
You can install them using pip:
pip install matplotlib numpy pandasNetMon can be used to visualize the training history and data distributions during the training of neural networks. Below is an example of how to use NetMon in your Python code:
from netmon import NeuralNetworkVisualizer
# Create a NetMon visualizer instance
visualizer = NeuralNetworkVisualizer()
# Simulate training process and update the history
for epoch in range(1, 11):
loss = 0.1 * epoch # Replace with actual loss values
accuracy = 0.7 + 0.05 * epoch # Replace with actual accuracy values
visualizer.update_training_history(epoch, loss, accuracy)
# Plot training history
visualizer.plot_training_history()
# Simulate data distribution
data = np.random.randn(1000) # Replace with actual data
visualizer.plot_data_distribution(data, title='Data Distribution')Visualize training history (loss and accuracy) over epochs. Plot data distribution for analysis. Easy integration with your neural network training code.
This project is licensed under the MIT License.