fix doc

docs/en/tutorials/quick_start_en.md

@@ -5,7 +5,7 @@ Based on the flowers102 dataset, it takes only 30 mins to experience PaddleClas,
 
 ## Preparation
 
-* enter insatallation dir
+* Enter insatallation dir.
 
 ```
 cd path_to_PaddleClas
@@ -16,7 +16,7 @@ cd path_to_PaddleClas
 ```shell
 cd dataset/flowers102
 # If you want to download from the brower, you can copy the link, visit it
-# in the browser, download and then commpress.
+# in the browser, download and then decommpress.
 wget https://paddle-imagenet-models-name.bj.bcebos.com/data/flowers102.zip
 unzip flowers102.zip
 ```
@@ -49,13 +49,25 @@ cd ../
 ## Training
 
 * All experiments are running on the NVIDIA® Tesla® V100 single card.
+* First of all, use the following command to set visible device.
+
+If you use mac or linux, you can use the following command:
+
+```shell
+export CUDA_VISIBLE_DEVICES=0
+```
+
+* If you use windows, you can use the following command.
+
+```shell
+set CUDA_VISIBLE_DEVICES=0
+```
 
 ### Train from scratch
 
 * Train ResNet50_vd
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd.yaml
 ```
 
@@ -66,10 +78,9 @@ The validation `Top1 Acc` curve is shown below.
 
 ### Finetune - ResNet50_vd pretrained model (Acc 79.12\%)
 
-* finetune ResNet50_vd_ model pretrained on the 1000-class Imagenet dataset
+* Finetune ResNet50_vd model pretrained on the 1000-class Imagenet dataset
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_finetune.yaml
 ```
 
@@ -120,11 +131,10 @@ pretrained_model: "./pretrained/ResNet50_vd_ssld_pretrained"
 Tringing script
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_ssld_finetune.yaml
 ```
 
-Compare with finetune on the 79.12% pretrained model, it improve by 0.9% to 95%.
+Compare with finetune on the 79.12% pretrained model, it improve by 0.98\% to 95\%.
 
 
 ### More architecture - MobileNetV3
@@ -132,7 +142,6 @@ Compare with finetune on the 79.12% pretrained model, it improve by 0.9% to 95%.
 Training script
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/MobileNetV3_large_x1_0_finetune.yaml
 ```
 
@@ -146,7 +155,6 @@ Data augmentation works when training data is small.
 Training script
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_ssld_random_erasing_finetune.yaml
 ```
 
@@ -180,7 +188,6 @@ TRAIN:
 Final training script
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/R50_vd_distill_MV3_large_x1_0.yaml
 ```
 

docs/zh_CN/tutorials/quick_start.md

@@ -47,6 +47,19 @@ cd ../
 ### 2.2 环境说明
 
 * 下面所有的训练过程均在`单卡V100`机器上运行。
+* 首先需要设置可用的显卡设备id
+
+如果使用mac或者linux，可以使用下面的命令进行设置。
+
+```shell
+export CUDA_VISIBLE_DEVICES=0
+```
+
+* 如果使用windows，可以使用下面的命令进行设置。
+
+```shell
+set CUDA_VISIBLE_DEVICES=0
+```
 
 
 ## 三、模型训练
@@ -56,7 +69,6 @@ cd ../
 * 基于ResNet50_vd模型，训练脚本如下所示。
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd.yaml
 ```
 
@@ -70,7 +82,6 @@ python3 tools/train.py -c ./configs/quick_start/ResNet50_vd.yaml
 * 基于ImageNet1k分类预训练模型进行微调，训练脚本如下所示。
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_finetune.yaml
 ```
 
@@ -80,7 +91,6 @@ python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_finetune.yaml
 
 使用训练完的预训练模型对图片`docs/images/quick_start/flowers102/image_06739.jpg`进行预测，预测命令为
 
-
 ```shell
 python3 tools/infer/infer.py \
     -i docs/images/quick_start/flowers102/image_06739.jpg \
@@ -119,11 +129,10 @@ pretrained_model: "./pretrained/ResNet50_vd_ssld_pretrained"
 训练脚本如下。
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_ssld_finetune.yaml
 ```
 
-最终flowers102验证集上精度指标为0.95，相对于79.12\%预训练模型的微调结构，新数据集指标可以再次提升0.9\%。
+最终flowers102验证集上精度指标为0.95，相对于79.12\%预训练模型的微调结构，新数据集指标可以再次提升0.98\%。
 
 
 ### 3.4 尝试更多的模型结构-MobileNetV3
@@ -131,7 +140,6 @@ python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_ssld_finetune.yam
 训练脚本如下所示。
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/MobileNetV3_large_x1_0_finetune.yaml
 ```
 
@@ -144,7 +152,6 @@ python3 tools/train.py -c ./configs/quick_start/MobileNetV3_large_x1_0_finetu
 
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/ResNet50_vd_ssld_random_erasing_finetune.yaml
 ```
 
@@ -180,7 +187,6 @@ TRAIN:
 最终的训练脚本如下所示。
 
 ```shell
-export CUDA_VISIBLE_DEVICES=0
 python3 tools/train.py -c ./configs/quick_start/R50_vd_distill_MV3_large_x1_0.yaml
 ```