A Framework for Deep Learning Research in PyTorch.
Project description
Fandak: فندک
Ideas
- Config files:
- One idea is to have config files (python files) with a specific signature that implements a specific function.
- These files should be able to be loaded dynamically.
- These files should be able to be saved in snapshot directories for resuming.
- But it is confusing for me how to manage the dependencies for these config files.
- I think 3 config files are needed: Train Dataset, Model, Training
- We can have one more config file for: Evaluation.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
fandak-0.0.1a1.dev3.tar.gz
(6.2 kB
view hashes)
Built Distribution
Close
Hashes for fandak-0.0.1a1.dev3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 594821b0495a2942ae7098e2cbf09b740979c95d6b8cde3ea25f969d20852183 |
|
MD5 | 32e46bd6dcc407fc3b50c5801fcb04db |
|
BLAKE2b-256 | 4ff4feb349903399b2d8be994882b82b0075db5ddd696aa00562997f87cb3bc4 |