iicd/autodl-projects
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update README

Browse Source
This commit is contained in:
D-X-Y
2019-09-05 15:15:04 +08:00
parent 9ad1125262
commit 1dfa4e1b46
3 changed files with 4 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
paddlepaddle/README.md
View File

@@ -1,7 +1,7 @@
# Image Classification based on NAS-Searched Models # Image Classification based on NAS-Searched Models
This directory contains 10 image classification models. This directory contains 10 image classification models.
Nine of them are automatically searched models from different Neural Architecture Search (NAS) algorithms. The other is the residual network. Nine of them are automatically searched models using different Neural Architecture Search (NAS) algorithms, and the other is the residual network.
We provide codes and scripts to train these models on both CIFAR-10 and CIFAR-100. We provide codes and scripts to train these models on both CIFAR-10 and CIFAR-100.
We use the standard data augmentation, i.e., random crop, random flip, and normalization. We use the standard data augmentation, i.e., random crop, random flip, and normalization.
@@ -18,6 +18,7 @@ We use the standard data augmentation, i.e., random crop, random flip, and norma
This project has the following requirements: This project has the following requirements:
- Python = 3.6 - Python = 3.6
- PadddlePaddle Fluid >= v0.15.0 - PadddlePaddle Fluid >= v0.15.0
- numpy, tarfile, cPickle, PIL
### Data Preparation ### Data Preparation
@@ -29,7 +30,7 @@ After data preparation, there should be two files `${TORCH_HOME}/cifar.python/ci
### Training Models ### Training Models
After setting up the environment and preparing the data, one can train the model. The main function entrance is `train_cifar.py`. We also provide some scripts for easy usage. After setting up the environment and preparing the data, you can train the model. The main function entrance is `train_cifar.py`. We also provide some scripts for easy usage.
``` ```
bash ./scripts/base-train.sh 0 cifar-10 ResNet110 bash ./scripts/base-train.sh 0 cifar-10 ResNet110
bash ./scripts/train-nas.sh 0 cifar-10 GDAS_V1 bash ./scripts/train-nas.sh 0 cifar-10 GDAS_V1
@@ -41,7 +42,7 @@ bash ./scripts/train-nas.sh 0 cifar-10 AmoebaNet
bash ./scripts/train-nas.sh 0 cifar-10 PNASNet bash ./scripts/train-nas.sh 0 cifar-10 PNASNet
bash ./scripts/train-nas.sh 0 cifar-100 SETN bash ./scripts/train-nas.sh 0 cifar-100 SETN
``` ```
The first argument is the GPU-ID to train your program, the second argument is the dataset name, and the last one is the model name. The first argument is the GPU-ID to train your program, the second argument is the dataset name (`cifar-10` or `cifar-100`), and the last one is the model name.
Please use `./scripts/base-train.sh` for ResNet and use `./scripts/train-nas.sh` for NAS-searched models. Please use `./scripts/base-train.sh` for ResNet and use `./scripts/train-nas.sh` for NAS-searched models.

3
paddlepaddle/lib/utils/meter.py
View File

@@ -1,6 +1,3 @@
##################################################
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2019 #
##################################################
import time, sys import time, sys
import numpy as np import numpy as np

6
paddlepaddle/lib/utils/time_utils.py
View File

@@ -1,9 +1,3 @@
# Copyright (c) Facebook, Inc. and its affiliates.
# All rights reserved.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
#
import time, sys import time, sys
import numpy as np import numpy as np