AI

Retrain_Inception_Resenet_V2 Tutorial

tensorflow平台模型網址 https://tfhub.dev/s?module-type=image-classification

--建立虛擬環境:
conda create –n tf python=3.6

--安裝tensorflow
pip install tensorflow tensorflow_hub

--1.切換工作目錄
cd /d C:\Users\n000022785\Downloads\TL activate tf

--2.訓練模型語法
python retrain.py --image_dir flower_photos
[ok] python retrain_IRV2.py --image_dir FHI_RT
[ok] python retrain_IRV2.py --image_dir FHIRTTrain
[ok] python retrain.py --image_dir FHI_RT

--3.測試影像語法
--單張
python label_image.py --graph=c:\tmp\output_graph.pb --labels=c:\tmp\output_labels.txt --input_layer=Placeholder --output_layer=final_result --image=RT_test/51.jpg

--多張
python label_image_loop.py --graph=c:\tmp\output_graph.pb --labels=c:\tmp\output_labels.txt --input_layer=Placeholder --output_layer=final_result --image=RT_test
python fhi_test.py --graph=c:\tmp\output_graph.pb --labels=c:\tmp\output_labels.txt --input_layer=Placeholder --output_layer=final_result --image=test1

--imagenet/inception_resnet_v2/classification
https://tfhub.dev/google/imagenet/inception_resnet_v2/classification/1

GAN Reference

• https://towardsdatascience.com/training-gans-using-google-colaboratory-f91d4e6f61fe
• https://github.com/rajaswa/Digit-Generation-by-GAN/blob/master/main.ipynb