XiangLi1999 Diffusion-LM issues

XiangLi1999 / Diffusion-LM

Diffusion-LM

Apache License 2.0

1.05k stars 135 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Is there any clear protocol for re-implementing this paper?

#73 Dandelionym opened 2 months ago
0
Difficulty in running code

#72 ajtheb opened 4 months ago
0
Seq2Seq tasks with Diffusion LM

#71 chiral-carbon opened 8 months ago
0
How to Execute the Semantic Content Subtask with infill.py

#70 pluto12x opened 10 months ago
0
Issue while generating controllable text generation

#69 heychhavi opened 11 months ago
0
Updated codes

#68 heychhavi opened 11 months ago
0
E2E training procedure

#67 elephantmipt opened 1 year ago
0
Questions about the NLL loss

#66 AlonzoLeeeooo opened 1 year ago
0
The relevant code that caused the error is in the Controllable Text Generation section, after the model trained for 6 epochs and started evaluating, it raised a KeyError: 'eval_loss'

#65 Markkk111 opened 1 year ago
2
The difference between this code and the paper "IDDPM" in the run_loop function in train_util.py.

#64 xyz321123 opened 1 year ago
0
About the tT_loss

#63 zzc681 opened 1 year ago
2
Which classifier to use in custom_trainer.py for controllable generation?

#62 CSerxy opened 1 year ago
0
error when runing：Exception in thread Thread-4:·······ValueError: signal number 32 out of range

#61 Markkk111 closed 1 year ago
0
Baseline reproduction

#60 YANI-ALT opened 1 year ago
0
Reproducing Table 5: Sentence Infilling - CIDEr / BLEU-4 metrics

#59 yair-schiff opened 1 year ago
1
Fail to load GPT2 pretrained model for attribute controled generation

#58 Rongzhi-Dong opened 1 year ago
0
Error when running training script on Google Colab

#57 ostov-larion closed 1 year ago
2
Why not directly use Emb(W) as X_0?

#56 leekum2018 opened 1 year ago
2
Questions about the result of success rate of PPLM?

#55 jiqimaoke closed 1 year ago
2
Separate weights for word embedding and lm-head?

#54 zeyun-zhong opened 1 year ago
0
Training on A100

#53 mathematiguy opened 1 year ago
0
I wander where to find the model in the predictability

#52 Layort opened 1 year ago
1
some problems on reproducing the results

#51 arealgoodname opened 1 year ago
0
Training Cost due to the EMA mechanism

#50 Lancelot39 opened 1 year ago
0
Do we need to scale word embeddings to [-1, 1]?

#49 tj-zhu opened 1 year ago
3
Why only padding tokens are generated after a period of training, but no words?

#48 greens007 opened 1 year ago
0
Losses for E2E Training

#47 zanussbaum closed 1 year ago
2
How could I perform semantic experiment in e2e dataset?

#46 WangCan1178 opened 1 year ago
2
Where is the mask indicator to calculate the decoder loss?

#45 smiles724 closed 1 year ago
2
Why model.model.module instead of model.model?

#44 smiles724 closed 2 years ago
0
Where is the mbr.py file?

#43 smiles724 closed 1 year ago
1
Is the embedding model trainable during the training process?

#42 smiles724 opened 2 years ago
3
How to deal with sequences with different lengths?

#41 sunhaozhepy opened 2 years ago
1
What if DiffusionLM is initialized with BERT?

#40 Hzfinfdu opened 2 years ago
2
How did you derive your sampling algo?

#39 jzhang38 opened 2 years ago
4
top_p parameter and scaling of timesteps

#38 rabeeh-karimi opened 2 years ago
1
How to train a new diffusion model & classifer with different diff_steps or embedding dimension?

#37 ChorlingLau opened 2 years ago
2
Infilling and text generation in specific contexts.

#36 radi-cho opened 2 years ago
5
Wandb log or Codalab log

#35 ShilongYuan opened 2 years ago
1
about {path-to-diffusion-lm}

#34 lcy5058 opened 2 years ago
8
decoding diffusion-lm and Controllable text generation

#33 JamesL404 opened 2 years ago
7
Error message when trying to train the model

#32 JamesL404 closed 2 years ago
3
ModuleNotFoundError: No module named 'mpi4py'

#31 cameronalonso2 closed 2 years ago
1
Is it possible to pre-train a diffusion language model on Wikipedia texts?

#30 RefluxNing opened 2 years ago
4
How to control the length

#29 gwang-kim opened 2 years ago
1
Would you please explain all the experiment names in your code?

#28 beyondguo opened 2 years ago
3
The effect of "logp_term"

#27 lgs00 opened 2 years ago
1
train the diffusion model for sentence infilling

#26 lwmlyy closed 2 years ago
1
Are these normal results？

#25 ChorlingLau opened 2 years ago
13
Create LICENSE

#24 XiangLi1999 closed 2 years ago
0