example for model training, saving and loading weights

[Kar1s]

hello,

Thank you for the nice framework.
Is it possible to save and load the weights of a model? It would also be helpful if you could provide another example regarding training with multiple sample points and labels.

[Rahul29999]

Hi @Kar1s,

I've worked on an end-to-end image classification project using PyTorch that aligns with the goals of this issue. It includes model training with multiple labeled data points, saving the trained weights, and loading them again for inference.

I’m sharing a minimal PyTorch example that demonstrates:

• Training a CNN model on synthetic image data
• Saving the trained model’s weights
• Loading the saved weights for inference

This example uses torch.save() and load_state_dict() to meet the requirements mentioned in Issue #92.

To make it easier to test and explore, I’ve attached the script as a .txt (train_save_load_example.txt).
Additionally, the full project is available here:

GitHub Repository: https://github.com/Rahul29999/-Animal-Classification-CNN-PyTorch-

Kaggle Notebook: https://www.kaggle.com/code/oops26/animal-classification-cnn-pytorch

Please let me know if a notebook version is preferred or if any modifications are needed. Glad to contribute and collaborate.

Thanks and best regards,
Rahul Kumar Sharma
B.Tech, IIT Dhanbad

Attached file : train_save_load_example.txt

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity in the past 90 days.

Note: This repository is archived and in maintenance mode. We are only accepting:

• Security fixes
• Critical bug fixes
• Documentation corrections

If this issue is still relevant, please comment to keep it open. Otherwise, it will be closed in 14 days.

Thank you for your contributions to LNN!