issues
search
eole-nlp
/
eole
Open language modeling toolkit based on PyTorch
https://eole-nlp.github.io/eole
MIT License
62
stars
12
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
fix maybe_retranslate when number of newline does not match
#150
vince62s
closed
1 day ago
0
flash_attn_func does not support padding mask maybe we need to drop a…
#149
vince62s
closed
1 day ago
0
fix missing pad change
#148
vince62s
closed
1 day ago
0
Model Validator Recipe
#146
francoishernandez
opened
3 days ago
0
CUDA Indexing Assertion Failure in Transformer Decoder
#145
HURIMOZ
opened
3 days ago
5
Torch 2.4.1 / 2.5 support
#144
francoishernandez
opened
5 days ago
0
Some minor fixes
#143
francoishernandez
closed
5 days ago
0
fix lora lm head
#142
vince62s
closed
1 day ago
0
fix autocast at scoring when doing AMP
#141
vince62s
closed
3 weeks ago
0
fix head dim in rope
#140
vince62s
closed
3 weeks ago
0
[BUG] subdirs cleanup
#139
vince62s
closed
5 days ago
0
fix #136. Updated eole/bin/model/average_models.py to work with safetensors model format.
#137
isanvicente
closed
3 weeks ago
1
eole model average broken?
#136
isanvicente
closed
3 weeks ago
5
add estimator in greedy inference
#135
vince62s
closed
4 weeks ago
0
Some QOL config/saving improvements
#134
francoishernandez
closed
4 weeks ago
0
enable pure bf16 training
#133
vince62s
closed
1 month ago
0
fixes #131, module 'eole.utils' has no attribute 'distributed' error when training multi-gpu
#132
isanvicente
closed
4 weeks ago
4
Error when training WMT with multi gpu
#131
isanvicente
closed
4 weeks ago
2
train_from option available?
#130
HURIMOZ
closed
1 month ago
1
Update WMT17 recipe with working tokenization transforms examples
#129
francoishernandez
closed
1 month ago
1
Bleu score for Tensorboard
#128
HURIMOZ
closed
1 month ago
11
eole predict command
#126
HURIMOZ
closed
1 month ago
25
Fix coverage penalty
#125
l-k-11235
closed
1 month ago
8
[docs] Upgrading docusaurus packages, should fix dependabot warnings
#124
francoishernandez
closed
1 month ago
0
Deduce share_decoder_embeddings from HF tie_word_embeddings flag
#123
francoishernandez
closed
1 month ago
0
[POC] Supporting standard tokenizers(HF, Mistral)
#122
francoishernandez
opened
1 month ago
0
Manage `share_decoder_embeddings` in `convert_HF`, misc fixes and improvements
#121
francoishernandez
closed
1 month ago
0
add estimator in decoder-only + clean code
#120
vince62s
closed
1 month ago
0
Fix coverage penalty (wu)
#119
l-k-11235
closed
1 month ago
2
Add comet scorer
#118
anthdr
opened
1 month ago
4
Missing key in safetensors checkpoint: generator.weight
#117
HURIMOZ
closed
1 month ago
3
restore all_reduce directly but with detach.clone first - fix #115
#116
vince62s
closed
1 month ago
0
fix training tensor parallel
#115
vince62s
closed
2 months ago
0
early_stopping not permitted?
#114
HURIMOZ
closed
1 month ago
1
Updated translator.py to handle updated special token logic when computing alignments
#113
dameikle
closed
2 months ago
0
clearer log
#112
vince62s
closed
2 months ago
0
[patch] get_transforms_cls after update_config_with_checkpoint BIS
#111
francoishernandez
closed
2 months ago
0
[patch] get_transforms_cls after update_config_with_checkpoint
#110
francoishernandez
closed
2 months ago
0
[patch] minor fixes for 0.0.2
#109
francoishernandez
closed
2 months ago
1
pre_word_vecs_enc attribute is not defined in the TrainConfig model
#107
HURIMOZ
opened
2 months ago
1
[patch] Update precision to compute_dtype in forgotten places
#106
francoishernandez
closed
2 months ago
0
[patch] Adapt some warning behaviours for reduced verbosity
#105
francoishernandez
closed
2 months ago
1
Remove duplicate sentencepiece requirement
#104
francoishernandez
closed
2 months ago
0
add wmt22 recipes with TowerInstruct and Llama3.1 LLMs
#103
vince62s
closed
2 months ago
0
Support mapped tokens eg: <im_start> ==> ｟im_start｠in inference.yaml …
#102
vince62s
closed
2 months ago
1
fix added tokens
#101
vince62s
closed
2 months ago
0
Fixdistrib
#100
vince62s
closed
2 months ago
0
Recipe stuck when predicting
#99
Kai-Piontek
closed
2 months ago
7
Initial support for Metal Performance Shaders (MPS)
#98
dameikle
closed
1 month ago
5
Sentencepiece
#97
VolodymyrB
closed
2 months ago
1
Next