paper: Densely Connected Convolutional Networks

This is the MXNet implement of DenseNet with pretrained model, therefor you can fine-tune in the pretrained model for your own dataset.

• Download pretrained models(pretrained in ImageNet dataset) from

These pretrained models are manually converted from https://github.com/shicai/DenseNet-Caffe ,put the pretrained model under /DenseNet/model/ file.

• I produce two ways of image data reading:

If you want to use .rec file to train your model:

• Change some configuration in run_train_rec.sh, for example: --epoch and --model are corresponding to the pretrained model, --data-train is your train .rec file, --save-result is the train result you want to save, --num-examples is the number of your training data, --save-name is the name of final model.
• Run

sh run_train_rec.sh

If you want to use .lst file and image to train your model:

• Change some configuration in run_train_lst.sh, for example: --epoch and --model are corresponding to the pretrained model, --data-train is your train .lst file, --image-train is your train image file, --save-result is the train result you want to save, --num-examples is the number of your training data, --save-name is the name of final model.
• Run

sh run_train_lst.sh