Zhaojing Luo

need to update documentation for distributed training, https://singa.apache.org/docs/dist-train/ e.g., adding explanations for conda package installation, optimizer selection, explanations for different data copies in different machines

Upload Release 4.0.0 Package to SVN

1

Upload Release 4.0.0 Package to SVN

All source files written by us MUST include the Apache license header: http://www.apache.org/legal/src-headers.html.

Switch between CPU and GPU devices for cnn example

3

For cifar_distributed_cnn and cnn examples, the model is by default run on the GPU. Need to enable it to run on CPU for machines with only CPUs

Intermediate information printing

3

Need to print intermediate information for cnn and cifar_distributed_cnn examples since it may take quite long for large models to finish one epoch of training

Save the downloaded datasets to local directory

2

In cnn and cifar_distributed_cnn examples, the downloaded files are in the /tmp directory, which may be lost after the machine is restarted.

Standardize the readme file, including dataset name, dataset link, download instructions, folder structure and the running command