iicd/xautodl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Upgrade API of NAS-Bench-201

Browse Source
This commit is contained in:
D-X-Y
2020-03-10 19:08:56 +11:00
parent c8f2a93ecf
commit d783193392
10 changed files with 623 additions and 178 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/ICCV-2019-SETN.md
View File

@@ -30,18 +30,14 @@ CUDA_VISIBLE_DEVICES=0 bash ./scripts/nas-infer-train.sh cifar100 SETN 96 -1
CUDA_VISIBLE_DEVICES=0,1,2,3 bash ./scripts/nas-infer-train.sh imagenet-1k SETN 256 -1
```
### Searching on the NASNet search space
Please use the following scripts to use SETN to search as in the original paper:
```
CUDA_VISIBLE_DEVICES=0 bash ./scripts-search/NASNet-space-search-by-SETN.sh cifar10 1 -1
```
### Searching on the NAS-Bench-201 search space
The searching codes of SETN on a small search space (NAS-Bench-201).
```
CUDA_VISIBLE_DEVICES=0 bash ./scripts-search/algos/SETN.sh cifar10 1 -1
```
**Searching on the NASNet search space** is not ready yet.
# Citation

7
docs/NAS-Bench-201.md
View File

@@ -21,9 +21,12 @@ You can simply type `pip install nas-bench-201` to install our api. Please see s
The benchmark file of NAS-Bench-201 can be downloaded from [Google Drive](https://drive.google.com/open?id=1SKW0Cu0u8-gb18zDpaAGi0f74UdXeGKs) or [Baidu-Wangpan (code:6u5d)](https://pan.baidu.com/s/1CiaNH6C12zuZf7q-Ilm09w).
You can move it to anywhere you want and send its path to our API for initialization.
- [2020.02.25] v1.0: `NAS-Bench-201-v1_0-e61699.pth`, where `e61699` is the last six digits for this file. It contains all information except for the trained weights of each trial.
- [2020.02.25] v1.0: The full data of each architecture can be download from [Google Drive](https://drive.google.com/open?id=1X2i-JXaElsnVLuGgM4tP-yNwtsspXgdQ) (about 226GB). This compressed folder has 15625 files containing the the trained weights.
- [2020.02.25] v1.0: The full data of each architecture can be download from [
NAS-BENCH-201-4-v1.0-archive.tar](https://drive.google.com/open?id=1X2i-JXaElsnVLuGgM4tP-yNwtsspXgdQ) (about 226GB). This compressed folder has 15625 files containing the the trained weights.
- [2020.02.25] v1.0: Checkpoints for 3 runs of each baseline NAS algorithm are provided in [Google Drive](https://drive.google.com/open?id=1eAgLZQAViP3r6dA0_ZOOGG9zPLXhGwXi).
- [2020.03.08] v2.0: coming soon (results of two set of hyper-parameters avaliable on all three datasets)
- [2020.03.09] v1.2: More robust API with more functions and descriptions
- [2020.04.01] v2.0: coming soon (results of two set of hyper-parameters avaliable on all three datasets)
The training and evaluation data used in NAS-Bench-201 can be downloaded from [Google Drive](https://drive.google.com/open?id=1L0Lzq8rWpZLPfiQGd6QR8q5xLV88emU7) or [Baidu-Wangpan (code:4fg7)](https://pan.baidu.com/s/1XAzavPKq3zcat1yBA1L2tQ).
It is recommended to put these data into `$TORCH_HOME` (`~/.torch/` by default). If you want to generate NAS-Bench-201 or similar NAS datasets or training models by yourself, you need these data.