Accessing values at internal nodes and carry over of value matrix values

[ritwickchaudhry]

I really liked the DKVMN model and wished to extend it further. I have read the code in detail and had some doubts:

1. Does the value matrix re-initialize every batch or are the values carried over?
2. Can the values at an internal node (Let's say the last fully connected layer) be
   I would be really grateful if you could help me in answering this

[jennyzhang0215]

1. Yes, the value matrix needs to re-initialize every batch.
2. I'm sorry I don't quite understand your question.

[ritwickchaudhry]

I am sorry I didn't complete the question. I wanted to ask if we can access the values of the internal variables in the network at training / testing time. For example the values in the 'memory_value" i.e it the value matrix in memory.py

Variable in code

I can access the values of the params that are saved/loaded by save_checkpoint but not other internal values

[sxjscience]

@ritwickchaudhry You may refer to the discussion here. apache/mxnet#4805