Densenet_multi_gpu_tfrecords

Densenet in this reposite, is based on the code at "https://github.com/LaurentMazare/deep-models/tree/master/densenet" by LaurentMazare, using tfrecords format data and either single cpu or multiple gpus if possible.

Differences

Some differences when compared with the design in the original paper (https://arxiv.org/abs/1608.06993):

• The number of layers in each block has been set to 6 or 12 for small memory.

Dataset

• The used dataset for test is flowers dataset. The images are converted into standard tfrecord dataset.

• You can choose any data with tfrecord datasets, only to modify the function: read_and_decode() and inputs()

• Input image size: 224*224*3

To do list

• Add the test step in the training.

• Memory-efficient version

• Update the ugly code