corenet icon indicating copy to clipboard operation
corenet copied to clipboard

CoreNet: A library for training deep neural networks

Results 39 corenet issues
Sort by recently updated
recently updated
newest added

In the Model Implementations section of the Directory Structure, I've made the following correction: | **From** | **To** | |----------|--------| | moels | models | In OpenELM Documentation, I've made...

hello, where is checkpoints?

Hey hey! - I'm VB, I work on the open source team at Hugging Face. Massive congratulations on the OpenELM release, it's quite refreshing to see such a brilliant open...

Fix snippet of python code at open_elm README. Before `PYTHONPATH=. mlx_examples/open_elm/inference.py` After `PYTHONPATH=. python3 mlx_examples/open_elm/inference.py`

via `typos --format brief --write-changes corenet` (via [typos-cli](https://github.com/crate-ci/typos)) Some manual (dictornaries -> dictionaries, overlayed -> overlay)

via `typos --format brief --write-changes` (via [typos-cli](https://github.com/crate-ci/typos))

reproducability -> reproducibility

## Example ``` opts = argparse.Namespace(**{"model.freeze_modules": "conv1"}) model = nn.Sequential( OrderedDict([ ('conv1', nn.Conv2d(20,64,5)), ('conv2', nn.Conv2d(20,64,5)) ]) ) print(freeze_modules_based_on_opts(opts, model)) ``` ![example_result](https://github.com/apple/corenet/assets/56601011/6d47bc00-5bc9-4fbc-a874-faa1e8e8a130)

I've made nested module selection based on the way the CSS children selector works. By using '>' we can now select nested modules. Example: ``` opts = argparse.Namespace(**{"model.freeze_modules": "model1>ins_model"}) inside_model2...

enhancement