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

update README

Browse Source
This commit is contained in:
D-X-Y
2019-09-28 20:18:18 +10:00
parent 180702ab8e
commit f8f3f382e0
18 changed files with 9 additions and 779 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
README.md
View File

@@ -50,7 +50,7 @@ Highlight: we equip one-shot NAS with an architecture sampler and train network
<img src="https://d-x-y.github.com/resources/paper-icon/ICCV-2019-SETN.png" width="450">
### Usage
Train the searched SETN-searched CNN on CIFAR-10, CIFAR-100, and ImageNet.
Please use the following scripts to train the searched SETN-searched CNN on CIFAR-10, CIFAR-100, and ImageNet.
```
CUDA_VISIBLE_DEVICES=0 bash ./scripts/nas-infer-train.sh cifar10 SETN 96 -1
CUDA_VISIBLE_DEVICES=0 bash ./scripts/nas-infer-train.sh cifar100 SETN 96 -1
@@ -64,12 +64,13 @@ Searching codes come soon!
We proposed a gradient-based searching algorithm using differentiable architecture sampling (improving DARTS with Gumbel-softmax sampling).
<img src="https://d-x-y.github.com/resources/paper-icon/CVPR-2019-GDAS.png" width="350">
<img src="https://d-x-y.github.com/resources/paper-icon/CVPR-2019-GDAS.png" width="300">
The old version is located at [`others/GDAS`](https://github.com/D-X-Y/NAS-Projects/tree/master/others/GDAS) and a paddlepaddle implementation is locate at [`others/paddlepaddle`](https://github.com/D-X-Y/NAS-Projects/tree/master/others/paddlepaddle).
### Usage
Train the searched GDAS-searched CNN on CIFAR-10, CIFAR-100, and ImageNet.
Please use the following scripts to train the searched GDAS-searched CNN on CIFAR-10, CIFAR-100, and ImageNet.
```
CUDA_VISIBLE_DEVICES=0 bash ./scripts/nas-infer-train.sh cifar10 GDAS_V1 96 -1
CUDA_VISIBLE_DEVICES=0 bash ./scripts/nas-infer-train.sh cifar100 GDAS_V1 96 -1