Fix a few issues in EfficientNet introduced by API change and rephrase documentation #1926
Conversation
README.md
Outdated
| @@ -88,6 +88,7 @@ Within the following table, we summarized the current NNI capabilities, we are g | |||
| <li><a href="docs/en_US/TrialExample/GbdtExample.md">Auto-gbdt</a></li> | |||
| <li><a href="docs/en_US/TrialExample/Cifar10Examples.md">Cifar10-pytorch</li></a> | |||
| <li><a href="docs/en_US/TrialExample/SklearnExamples.md">Scikit-learn</a></li> | |||
| <li><a href="examples/trials/efficientnet">EfficientNet</a></li> | |||
There was a problem hiding this comment.
Why not put the doc under the "docs/en_US/TrialExample/"?
There was a problem hiding this comment.
If the docs are to be put under TrialExample, they are intended to be read on readthedocs, which means every link on README should be a link to readthedocs website, instead of a link to a markdown on Github. Refactoring involves many changes and should be open to discussion.
There was a problem hiding this comment.
If you concern that you cannot support markdown and readthedocs at the same time, you could put in the path but do not link to readthedocs in rst file.
The motivation that we build a doc folder is that we want to organize all the doc in this folder.
There was a problem hiding this comment.
@ultmaster please put it under docs/
|
|
||
| ## Instructions | ||
|
|
||
| 1. Set your working directory here in this directory. | ||
| 2. Run `git clone https://github.com/ultmaster/EfficientNet-PyTorch` to clone this modified version of [EfficientNet-PyTorch](https://github.com/lukemelas/EfficientNet-PyTorch). The modifications were done to adhere to the original [Tensorflow version](https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet) as close as possible (including EMA, label smoothing and etc.); also added are the part which gets parameters from tuner and reports intermediate/final results. Clone it into `EfficientNet-PyTorch`; the files like `main.py`, `train_imagenet.sh` will appear inside, as specified in the configuration files. | ||
| 3. Run `nnictl create --config config_net.yml` to find the best EfficientNet-B1. Adjust the training service (PAI/local/remote), batch size in the config files according to the environment. | ||
| 3. Run `nnictl create --config config_local.yml` (use `config_pai.yml` for PAI) to find the best EfficientNet-B1. Adjust the training service (PAI/local/remote), batch size in the config files according to the environment. |
There was a problem hiding this comment.
I am sure that we should use PAI or OpenPAI?
No description provided.