update TODO

Pabanjyoti · Mar 26, 2020 · d053a61 · d053a61
1 parent daef13f
commit d053a61
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 2 deletions.
diff --git a/README.md b/README.md
@@ -2,10 +2,18 @@
 
 > NTU-AI6127 NLP Final Project
 
-### Ref
+### Reference
 
 * [a-PyTorch-Tutorial-to-Image-Captioning](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Image-Captioning)
 
+### TODO
+
+- [x] Baseline.
+- [ ]  Glove pre-trained word embedding
+- [ ] [Multi-Eva metrics](https://github.com/tylin/coco-caption)
+- [ ] [self-attentiion:Transformer](https://github.com/huggingface/transformers)
+- [ ] try [Multi layer RNN](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Image-Captioning/pull/79) or different attention (may discard)
+
 ### Requirements
 
 Recommend for `Conda` Env.

diff --git a/train.py b/train.py
@@ -242,7 +242,7 @@ def validate(args, val_loader, encoder, decoder, criterion):
                                            dropout=args.dropout)
             decoder_optimizer = torch.optim.Adam(params=filter(lambda p: p.requires_grad, decoder.parameters()),
                                                  lr=args.decoder_lr)
-        else:
+        elif args.decoder_mode == "TRANSFORMER":
             # TODO: Transformer
             pass