[SINGA-264] Extend the FeedForwardNet to accept multiple inputs

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Assignee: wangwei
• Reporter: wangwei
• Created: 18/Oct/16 13:43
• Updated: 11/Jan/17 06:46
• Resolved: 11/Jan/17 06:46

Description

The current implementation of the feed-forward net assumes that the input data comes from a single source, which does not work for nets with multiple input sources.

To enable multiple data sources and to be compatible with existing code, we can update the code by adding a check for the x field of `def train(x, y)`

if type(x) is dict:  # multi sources
    # feed the value x[k] to the input layer with name k
else:  # single source
   feed x to the first layer, i.e. the input layer

Similarly we can also process multiple output (label) tensors.