New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Research result validation process #24

Open

SteelStrider opened this issue Jun 6, 2018 · 1 comment

Assignees

Contributor

SteelStrider commented Jun 6, 2018

No description provided.

SteelStrider self-assigned this

Contributor Author

SteelStrider commented Jun 18, 2018

One of the working ideas is the independent formation of data for testing the weights of a trained model. For example, in the following form.
PyNode gets the task to train the model, performs it. -> result = file with trained weights

Other PyNode receives the task to perform validation

Gets the data for training
Randomly selects a certain percentage of the data for validation from the training batches
Conducts evaluate() on selected data
If the result has a small percentage of losses and sufficient accuracy
- considers work on training performed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment