facebookresearch XLM issues

facebookresearch / XLM

PyTorch original implementation of Cross-lingual Language Model Pretraining.

Other

2.87k stars 493 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Predict a masked word

#361 ebilal opened 9 months ago
0
how to save the entire model instead of just the model parameters

#360 vicky-yuan opened 1 year ago
0
bt_steps meaning

#359 rayendito closed 1 year ago
0
get-data-glue.sh 400 Bad Request

#358 t109368507 opened 1 year ago
0
Error in model, scaling only q matrix not qK.T dot product (qk.T/sqrt(dim_per_head))

#357 BenoitDalFerro opened 1 year ago
2
e

#356 jvamvas closed 1 year ago
0
How can I expand it to a new language which is Romanised? For example, Marathi Romanized?

#355 aayushkubb opened 1 year ago
0
confusion about `lm_head`'s size?

#354 tnq177 opened 1 year ago
2
Checkpoint for TLM objective

#353 xu1998hz opened 1 year ago
0
./get-data-para.sh

#352 Sakurakdx opened 1 year ago
3
[Question] Does XLM-R follows RoBERTa or XLM for MLM?

#351 mani-rai opened 2 years ago
0
How is sentence piece model trained in XLM-R?

#350 mani-rai opened 2 years ago
0
supervised machine translation

#349 yelinga opened 2 years ago
1
default params for PKM

#348 rabeehk opened 2 years ago
0
Question about parameters for further training of a preexisiting model?

#347 mcriggs opened 2 years ago
0
Fix typo for Sanskrit

#346 raghothams opened 2 years ago
2
Training data details for XLM-15 model

#345 somani-iitb opened 2 years ago
1
Generate multiple optimal results（beam search）

#344 Zhw098 opened 2 years ago
0
Error in Training

#343 sadanyh opened 2 years ago
3
Error when using the uploaded en-fr model for NMT (translate from English to French)

#342 Anwarvic opened 2 years ago
1
XLM LICENSE

#341 asm-cygu opened 2 years ago
0
Add memory to transformer

#340 Arij-Aladel opened 2 years ago
0
Getting Assertion error: How to use XLM for Unsupervised NMT of language pairs other than English-French, English-German and English-Romanian

#339 rashikumar01 opened 3 years ago
1
BOS is not used during training

#338 leo-liuzy opened 3 years ago
0
How to use XLM-R?

#337 leo-liuzy closed 3 years ago
3
Missing WikiExtractor.py file when running get-data-wiki.sh

#335 wayi1 closed 2 years ago
2
./get-data-glue.sh

#334 leaves-slient opened 3 years ago
1
Vocab size not match model input size

#333 moment-of-peace opened 3 years ago
1
finetune on GLUE task ends up with same probabality

#332 TingchenFu opened 3 years ago
2
Find subpackages in xlm setup

#331 jrapin closed 3 years ago
1
Fix typo in README

#330 mzaidi59 opened 3 years ago
1
Language model training data

#329 sbmaruf opened 3 years ago
0
Clarification regarding emb_dim parameter value used in the paper

#328 asolano opened 3 years ago
1
cannot find checkpoint to reload in multi-gpu pretraining

#327 colmantse closed 3 years ago
0
Questions about zh-en pre-training model

#326 hcd7434 opened 3 years ago
0
Difference between code and vocabulary

#325 Hannibal046 closed 3 years ago
0
Multiple GPU speedup

#324 asolano opened 3 years ago
0
[Experiment settings]: the total training steps and batch size

#323 KyGao opened 3 years ago
0
Fix compatibility issues with WikiExtractor

#322 machelreid opened 3 years ago
3
ApplyBPE get empty file

#321 chenQ1114 opened 3 years ago
1
How to fix the batch

#320 hcd7434 opened 3 years ago
0
Get data generates empty files

#319 srcarroll opened 3 years ago
3
generate embeddings missing/unexpected keys error

#318 sylyoung closed 3 years ago
1
decipher additions

#317 pmulcaire opened 3 years ago
1
bleu score and subprocess.Popen communicate method

#316 Tikquuss closed 3 years ago
0
System requirment for XML model

#315 ykkhan opened 3 years ago
3
How to train unsupervised MT Language Model without parallel data?

#314 sylyoung closed 3 years ago
1
Can we apply this methodology on Win10?

#313 ykkhan closed 3 years ago
0
XLM-R fine-tune in MLQA dataset

#312 ztl-35 opened 4 years ago
2
Why batch of same language?

#311 hwijeen opened 4 years ago
0