图像分类集成以下模型:ResNet18、ResNet34、ResNet50、ResNet101、ResNet152、VGG16、VGG19、InceptionV3、Xception、MobileNet、AlexNet、LeNet、ZF_Net、DenseNet、mnist_net、TSL16,在config.py里面选择使用哪种模型.
- VGG16
- VGG19
- InceptionV3
- Xception
- MobileNet
- AlexNet
- LeNet
- ZF_Net
- ResNet18
- ResNet34
- ResNet50
- ResNet101
- ResNet152
- DenseNet(dismissed this time)
- mnist_net
- TSL16
"training or testing dataset folder is:"
/path/classes1/cat*.jpg,
/path/classes2/dog*.jpg,
/path/classes3/people*.jpg,
/path/classes4/*.jpg,
- Attentions ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
- classes name must be contained in folder name
My environment is based on
- ubuntu16
- cuda8 (cuda9.0)
- tensorflow_gpu1.4 (tensorflow_gpu1.10 )
- keras2.0.8
- numpy
- tqdm
- opencv-python
- scikit-learn
- pip3 install tensorflow_gpu==1.4
- pip3 install keras==2.0.8
- pip3 install numpy
- pip3 install tqdm
- pip3 install opencv-python
- pip3 install scikit-learn
- choose model and change parameter in config.py
- python3 mk_class_idx.py
- Train sigle model : python3 train.py modelName
- Train All model : run " sh trainAll.sh " to train all model (in ubuntu)
- Tensorboard : take LeNet as example, run " tensorboard --logdir=./checkpoints/LeNet " to watch training with tensorboard
- predict model: python3 predict.py modelName classesName
Author email: mymailwith163@163.com
I do not provide any trained model, and so, you should train your own model, if you get error: "rm: cannot remove 'checkpoints/ModelName/events*': No such file or directory", do not worry, It means you has not trained your model, and just go to enjoy training.