Commit Graph

414 Commits

Author SHA1 Message Date
lucidrains
410a6144e1 new einops is torch compile friendly 2023-10-18 15:45:09 -07:00
lucidrains
c6c3882dc1 fix all optional types in train config 2023-10-07 11:34:34 -07:00
Phil Wang
512b52bd78 1.15.2 2023-10-04 09:38:46 -07:00
Neil Kim Nielsen
147c156c8a Make TrackerLoadConfig optional (#306) 2023-10-04 09:38:30 -07:00
Phil Wang
40843bcc21 pydantic 2 2023-07-15 09:32:44 -07:00
Phil Wang
00e07b7d61 force einops 0.6.1 or greater and call allow_ops_in_compiled_graph 2023-04-20 14:08:52 -07:00
Phil Wang
0069857cf8 remove einops exts for better pytorch 2.0 compile compatibility 2023-04-20 07:05:29 -07:00
Phil Wang
580274be79 use .to(device) to avoid copy, within one_unet_in_gpu context 2023-03-07 12:41:55 -08:00
Phil Wang
848e8a480a always rederive the predicted noise from the clipped x0 for ddim + predict noise objective 2023-03-05 10:45:44 -08:00
Phil Wang
cc58f75474 bump to newer package of clip-anytorch that allows for text encodings < maximum context length 2023-03-04 09:37:25 -08:00
Phil Wang
3b2cf7b0bc fix for self conditioning in diffusion prior network https://github.com/lucidrains/DALLE2-pytorch/issues/273 2023-02-11 17:18:40 -08:00
Phil Wang
984d62a373 default ddim sampling eta to 0 2022-12-23 13:23:09 -08:00
Phil Wang
683dd98b96 extra insurance in case eos id is not there 2022-12-15 10:54:21 -08:00
Phil Wang
067ac323da address https://github.com/lucidrains/DALLE2-pytorch/issues/266 2022-11-23 08:41:25 -08:00
zion
91c8d1ca13 bug fix cosine annealing optimizer in prior trainer (#262) 2022-11-11 12:15:13 -08:00
zion
7166ad6711 add open clip to train_config (#260)
add the ability to use open_clip in the train configs (useful for the new SOTA h/14 model)
2022-11-07 15:44:36 -08:00
Phil Wang
fbba0f9aaf bring in prediction of v objective, combining the findings from progressive distillation paper and imagen-video to the eventual extension of dalle2 to make-a-video 2022-10-28 18:21:07 -07:00
Romain Beaumont
9f37705d87 Add static graph param (#226)
* Add static graph param

* use static graph param
2022-10-25 19:31:29 +02:00
Phil Wang
c3df46e374 fix openclipadapter to be able to use latest open sourced sota model 2022-10-23 15:12:09 -07:00
Phil Wang
41fabf2922 fix a dtype conversion issue for the diffusion timesteps in the diffusion prior, thanks to @JiaHeng-DLUT 2022-10-19 09:26:06 -07:00
Heng Jia
5975e8222b Fix assert message (#253) 2022-10-18 08:50:59 -07:00
Phil Wang
c18c080128 fix for use with larger openai clip models by extracting dimension of last layernorm in clip 2022-09-29 09:09:47 -07:00
Phil Wang
d0c11b30b0 handle open clip adapter image size being a tuple 2022-09-19 10:27:14 -07:00
Phil Wang
0d82dff9c5 in ddim, noise should be predicted after x0 is maybe clipped, thanks to @lukovnikov for pointing this out in another repository 2022-09-01 09:40:47 -07:00
Phil Wang
8bbc956ff1 fix bug with misnamed variable in diffusion prior network 2022-08-31 17:19:05 -07:00
Phil Wang
6fb7e91343 fix ddim to use alpha_cumprod 2022-08-31 07:40:46 -07:00
Phil Wang
ba58ae0bf2 add two asserts to diffusion prior to ensure matching image embedding dimensions for clip, diffusion prior network, and what was set on diffusion prior 2022-08-28 10:11:37 -07:00
Phil Wang
1cc5d0afa7 upgrade to best downsample 2022-08-25 10:37:02 -07:00
Phil Wang
59fa101c4d fix classifier free guidance for diffusion prior, thanks to @jaykim9870 for spotting the issue 2022-08-23 08:29:01 -07:00
Aidan
cbaadb6931 Fixed issues with clip and deepspeed fp16
Also more more general compatibility fixes
2022-08-20 17:58:32 +00:00
Phil Wang
083508ff8e cast attention matrix back to original dtype pre-softmax in attention 2022-08-20 10:56:01 -07:00
Phil Wang
7762edd0ff make it work for @ethancohen123 2022-08-19 11:28:58 -07:00
Phil Wang
27f19ba7fa make sure diffusion prior trainer can operate with no warmup 2022-08-15 14:27:40 -07:00
Phil Wang
8f38339c2b give diffusion prior trainer cosine annealing lr too 2022-08-15 07:38:01 -07:00
Phil Wang
6b9b4b9e5e add cosine annealing lr schedule 2022-08-15 07:29:56 -07:00
Phil Wang
44e09d5a4d add weight standardization behind feature flag, which may potentially work well with group norm 2022-08-14 11:34:45 -07:00
Phil Wang
34806663e3 make it so diffusion prior p_sample_loop returns unnormalized image embeddings 2022-08-13 10:03:40 -07:00
Phil Wang
dc816b1b6e dry up some code around handling unet outputs with learned variance 2022-08-12 15:25:03 -07:00
Phil Wang
05192ffac4 fix self conditioning shape in diffusion prior 2022-08-12 12:30:03 -07:00
Phil Wang
9440411954 make self conditioning technique work with diffusion prior 2022-08-12 12:20:51 -07:00
Phil Wang
981d407792 comment 2022-08-12 11:41:23 -07:00
Phil Wang
7c5477b26d bet on the new self-conditioning technique out of geoffrey hintons group 2022-08-12 11:36:08 -07:00
Phil Wang
be3bb868bf add gradient checkpointing for all resnet blocks 2022-08-02 19:21:44 -07:00
Phil Wang
451de34871 enforce clip anytorch version 2022-07-30 10:07:55 -07:00
Phil Wang
f22e8c8741 make open clip available for use with dalle2 pytorch 2022-07-30 09:02:31 -07:00
Phil Wang
87432e93ad quick fix for linear attention 2022-07-29 13:17:12 -07:00
Phil Wang
d167378401 add cosine sim for self attention as well, as a setting 2022-07-29 12:48:20 -07:00
Phil Wang
2d67d5821e change up epsilon in layernorm the case of using fp16, thanks to @Veldrovive for figuring out this stabilizes training 2022-07-29 12:41:02 -07:00
Phil Wang
748c7fe7af allow for cosine sim cross attention, modify linear attention in attempt to resolve issue on fp16 2022-07-29 11:12:18 -07:00
Phil Wang
80046334ad make sure entire readme runs without errors 2022-07-28 10:17:43 -07:00