diff options
author | Gustaf Rydholm <gustaf.rydholm@gmail.com> | 2021-08-10 23:25:13 +0200 |
---|---|---|
committer | Gustaf Rydholm <gustaf.rydholm@gmail.com> | 2021-08-10 23:25:13 +0200 |
commit | 441b7484348953deb7c94150675d54583ef5a81a (patch) | |
tree | 6a3e537f98a5c34fe166db7f4c60552b1212b8f5 /README.md | |
parent | 82f4acabe24e5171c40afa2939a4777ba87bcc30 (diff) |
Update to config and logging in VQGAN
Diffstat (limited to 'README.md')
-rw-r--r-- | README.md | 6 |
1 files changed, 1 insertions, 5 deletions
@@ -27,11 +27,7 @@ python build-transitions --tokens iamdb_1kwp_tokens_1000.txt --lexicon iamdb_1kw (TODO: Not working atm, needed for GTN loss function) ## Todo -- [x] Efficient-net b0 + transformer decoder -- [x] Load everything with hydra, get it to work -- [x] Train network -- [ ] Weight init -- [ ] patchgan loss +- [ ] patchgan loss FIX THIS!! LOOK AT TAMING TRANSFORMER, MORE SPECIFICALLY SEND LAYER AND COMPUTE COEFFICIENT - [ ] Get VQVAE2 to work and not get loss NAN - [ ] Local attention for target sequence - [ ] Rotary embedding for target sequence |