dcslin commented on pull request #697:
URL: https://github.com/apache/singa/pull/697#issuecomment-631276040
> > TODO list compiled by @XJDKC with slight update:
> >
> > * [x] [rulin] 1. separate layers from autograd.py to form a new
file(`layer.py`)
> > _Alias added in `autograd.py` for backward compatibility_
> > * [x] [rulin] 2. rename the Operation to Operator
> > * [x] [rulin] 3. create alias for Operator and Tensor
> > * [x] 4. make Model(_renamed from module_) a subclass of Layer
> > * [ ] 5. compile / do_init: separate the initialzation and forward
propagation for each layers, create name of each layer
> > **Done for: Linear**
> > * [x] [rulin] 6. train_one_batch: with graph and without graph
> > * [ ] 7. save_states/load_states/set_states: recursive logic」
TODO1: return flatten dict in each layers
TODO2: add params into states
[ ] 8. give name (attribute) to `layers` in `model.compile`, and
Tensor params `Tensor.name = "l1:W"`
>
> added update
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
