Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- config.json +1435 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- model.onnx +3 -0
- model.onnx_data +3 -0
- special_tokens_map.json +20 -0
- tokenizer_config.json +228 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
model.onnx_data filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
|
@@ -0,0 +1,1435 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "xlm-mlm-100-1280",
|
| 3 |
+
"accumulate_gradients": 4,
|
| 4 |
+
"ae_steps": [],
|
| 5 |
+
"amp": 2,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"XLMWithLMHeadModel"
|
| 8 |
+
],
|
| 9 |
+
"asm": false,
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"batch_size": 16,
|
| 12 |
+
"beam_size": 1,
|
| 13 |
+
"bos_index": 0,
|
| 14 |
+
"bos_token_id": 0,
|
| 15 |
+
"bptt": 256,
|
| 16 |
+
"bt_src_langs": [],
|
| 17 |
+
"bt_steps": [],
|
| 18 |
+
"causal": false,
|
| 19 |
+
"clip_grad_norm": 1.0,
|
| 20 |
+
"clm_steps": [],
|
| 21 |
+
"command": "python /private/home/aconneau/workdir/xlm_17_100_big.3/2019_08_10_19_23_42/train.py --n_heads 16 --bt_steps '' --max_vocab 200000 --word_mask_keep_rand '0.8,0.1,0.1' --use_lang_emb false --data_path '/private/home/aconneau/projects/XLM/data/wiki/100/175k' --save_periodic 0 --max_len 200 --bptt 256 --ae_steps '' --fp16 true --share_inout_emb true --sinusoidal_embeddings false --word_shuffle 0 --tokens_per_batch '-1' --accumulate_gradients 4 --validation_metrics '_valid_en_mlm_ppl,_valid_mlm_ppl,_valid_zh_mlm_ppl' --attention_dropout '0.1' --split_data true --max_epoch 100000 --stopping_criterion '_valid_zh_mlm_ppl,25' --dump_path '/checkpoint/aconneau/dumped' --epoch_size 200000 --word_blank 0 --gelu_activation true --n_layers 16 --optimizer 'adam_inverse_sqrt,lr=0.00005,warmup_updates=30000,beta1=0.9,beta2=0.999,weight_decay=0.01,eps=0.000001' --mlm_steps 'en,es,fr,de,zh,ru,pt,it,ar,ja,id,tr,nl,pl,simple,fa,vi,sv,ko,he,ro,no,hi,uk,cs,fi,hu,th,da,ca,el,bg,sr,ms,bn,hr,sl,zh_yue,az,sk,eo,ta,sh,lt,et,ml,la,bs,sq,arz,af,ka,mr,eu,tl,ang,gl,nn,ur,kk,be,hy,te,lv,mk,zh_classical,als,is,wuu,my,sco,mn,ceb,ast,cy,kn,br,an,gu,bar,uz,lb,ne,si,war,jv,ga,zh_min_nan,oc,ku,sw,nds,ckb,ia,yi,fy,scn,gan,tt,am' --eval_bleu false --dropout '0.1' --mt_steps '' --batch_size 16 --word_dropout 0 --reload_model '/checkpoint/aconneau/dumped/xlm_17_100_240_big_model_upper.2/14884511/best-valid_zh_mlm_ppl.pth' --min_count 0 --amp 2 --group_by_size true --asm false --sample_alpha '0.5' --word_pred '0.15' --clip_grad_norm 1 --emb_dim 1280 --encoder_only true --lgs 'en-es-fr-de-zh-ru-pt-it-ar-ja-id-tr-nl-pl-simple-fa-vi-sv-ko-he-ro-no-hi-uk-cs-fi-hu-th-da-ca-el-bg-sr-ms-bn-hr-sl-zh_yue-az-sk-eo-ta-sh-lt-et-ml-la-bs-sq-arz-af-ka-mr-eu-tl-ang-gl-nn-ur-kk-be-hy-te-lv-mk-zh_classical-als-is-wuu-my-sco-mn-ceb-ast-cy-kn-br-an-gu-bar-uz-lb-ne-si-war-jv-ga-zh_min_nan-oc-ku-sw-nds-ckb-ia-yi-fy-scn-gan-tt-am' --clm_steps '' --exp_name 'xlm_17_100_big.3' --lg_sampling_factor '0.7' --eval_only false --exp_id 16656234 --master_port 11363 --exp_id \"16656234\"",
|
| 22 |
+
"context_size": 0,
|
| 23 |
+
"data_path": "/private/home/aconneau/projects/XLM/data/wiki/100/175k",
|
| 24 |
+
"debug": false,
|
| 25 |
+
"debug_slurm": false,
|
| 26 |
+
"debug_train": false,
|
| 27 |
+
"dropout": 0.1,
|
| 28 |
+
"dump_path": "/checkpoint/aconneau/dumped/xlm_17_100_big.3/16656234",
|
| 29 |
+
"emb_dim": 1280,
|
| 30 |
+
"embed_init_std": 0.02209708691207961,
|
| 31 |
+
"encoder_only": true,
|
| 32 |
+
"end_n_top": 5,
|
| 33 |
+
"eos_index": 1,
|
| 34 |
+
"epoch_size": 200000,
|
| 35 |
+
"eval_bleu": false,
|
| 36 |
+
"eval_only": false,
|
| 37 |
+
"exp_id": "16656234",
|
| 38 |
+
"exp_name": "xlm_17_100_big.3",
|
| 39 |
+
"fp16": true,
|
| 40 |
+
"gelu_activation": true,
|
| 41 |
+
"global_rank": 0,
|
| 42 |
+
"group_by_size": true,
|
| 43 |
+
"hyp_path": "/checkpoint/aconneau/dumped/xlm_17_100_big.3/16656234/hypotheses",
|
| 44 |
+
"id2lang": {
|
| 45 |
+
"0": "af",
|
| 46 |
+
"1": "als",
|
| 47 |
+
"10": "be",
|
| 48 |
+
"11": "bg",
|
| 49 |
+
"12": "bn",
|
| 50 |
+
"13": "br",
|
| 51 |
+
"14": "bs",
|
| 52 |
+
"15": "ca",
|
| 53 |
+
"16": "ceb",
|
| 54 |
+
"17": "ckb",
|
| 55 |
+
"18": "cs",
|
| 56 |
+
"19": "cy",
|
| 57 |
+
"2": "am",
|
| 58 |
+
"20": "da",
|
| 59 |
+
"21": "de",
|
| 60 |
+
"22": "el",
|
| 61 |
+
"23": "en",
|
| 62 |
+
"24": "eo",
|
| 63 |
+
"25": "es",
|
| 64 |
+
"26": "et",
|
| 65 |
+
"27": "eu",
|
| 66 |
+
"28": "fa",
|
| 67 |
+
"29": "fi",
|
| 68 |
+
"3": "an",
|
| 69 |
+
"30": "fr",
|
| 70 |
+
"31": "fy",
|
| 71 |
+
"32": "ga",
|
| 72 |
+
"33": "gan",
|
| 73 |
+
"34": "gl",
|
| 74 |
+
"35": "gu",
|
| 75 |
+
"36": "he",
|
| 76 |
+
"37": "hi",
|
| 77 |
+
"38": "hr",
|
| 78 |
+
"39": "hu",
|
| 79 |
+
"4": "ang",
|
| 80 |
+
"40": "hy",
|
| 81 |
+
"41": "ia",
|
| 82 |
+
"42": "id",
|
| 83 |
+
"43": "is",
|
| 84 |
+
"44": "it",
|
| 85 |
+
"45": "ja",
|
| 86 |
+
"46": "jv",
|
| 87 |
+
"47": "ka",
|
| 88 |
+
"48": "kk",
|
| 89 |
+
"49": "kn",
|
| 90 |
+
"5": "ar",
|
| 91 |
+
"50": "ko",
|
| 92 |
+
"51": "ku",
|
| 93 |
+
"52": "la",
|
| 94 |
+
"53": "lb",
|
| 95 |
+
"54": "lt",
|
| 96 |
+
"55": "lv",
|
| 97 |
+
"56": "mk",
|
| 98 |
+
"57": "ml",
|
| 99 |
+
"58": "mn",
|
| 100 |
+
"59": "mr",
|
| 101 |
+
"6": "arz",
|
| 102 |
+
"60": "ms",
|
| 103 |
+
"61": "my",
|
| 104 |
+
"62": "nds",
|
| 105 |
+
"63": "ne",
|
| 106 |
+
"64": "nl",
|
| 107 |
+
"65": "nn",
|
| 108 |
+
"66": "no",
|
| 109 |
+
"67": "oc",
|
| 110 |
+
"68": "pl",
|
| 111 |
+
"69": "pt",
|
| 112 |
+
"7": "ast",
|
| 113 |
+
"70": "ro",
|
| 114 |
+
"71": "ru",
|
| 115 |
+
"72": "scn",
|
| 116 |
+
"73": "sco",
|
| 117 |
+
"74": "sh",
|
| 118 |
+
"75": "si",
|
| 119 |
+
"76": "simple",
|
| 120 |
+
"77": "sk",
|
| 121 |
+
"78": "sl",
|
| 122 |
+
"79": "sq",
|
| 123 |
+
"8": "az",
|
| 124 |
+
"80": "sr",
|
| 125 |
+
"81": "sv",
|
| 126 |
+
"82": "sw",
|
| 127 |
+
"83": "ta",
|
| 128 |
+
"84": "te",
|
| 129 |
+
"85": "th",
|
| 130 |
+
"86": "tl",
|
| 131 |
+
"87": "tr",
|
| 132 |
+
"88": "tt",
|
| 133 |
+
"89": "uk",
|
| 134 |
+
"9": "bar",
|
| 135 |
+
"90": "ur",
|
| 136 |
+
"91": "uz",
|
| 137 |
+
"92": "vi",
|
| 138 |
+
"93": "war",
|
| 139 |
+
"94": "wuu",
|
| 140 |
+
"95": "yi",
|
| 141 |
+
"96": "zh",
|
| 142 |
+
"97": "zh_classical",
|
| 143 |
+
"98": "zh_min_nan",
|
| 144 |
+
"99": "zh_yue"
|
| 145 |
+
},
|
| 146 |
+
"init_std": 0.02,
|
| 147 |
+
"is_encoder": true,
|
| 148 |
+
"is_master": true,
|
| 149 |
+
"is_slurm_job": true,
|
| 150 |
+
"lambda_ae": 1.0,
|
| 151 |
+
"lambda_ae_config": null,
|
| 152 |
+
"lambda_bt": 1.0,
|
| 153 |
+
"lambda_bt_config": null,
|
| 154 |
+
"lambda_clm": 1.0,
|
| 155 |
+
"lambda_clm_config": null,
|
| 156 |
+
"lambda_mlm": 1.0,
|
| 157 |
+
"lambda_mlm_config": null,
|
| 158 |
+
"lambda_mt": 1.0,
|
| 159 |
+
"lambda_mt_config": null,
|
| 160 |
+
"lambda_pc": 1.0,
|
| 161 |
+
"lambda_pc_config": null,
|
| 162 |
+
"lang2id": {
|
| 163 |
+
"af": 0,
|
| 164 |
+
"als": 1,
|
| 165 |
+
"am": 2,
|
| 166 |
+
"an": 3,
|
| 167 |
+
"ang": 4,
|
| 168 |
+
"ar": 5,
|
| 169 |
+
"arz": 6,
|
| 170 |
+
"ast": 7,
|
| 171 |
+
"az": 8,
|
| 172 |
+
"bar": 9,
|
| 173 |
+
"be": 10,
|
| 174 |
+
"bg": 11,
|
| 175 |
+
"bn": 12,
|
| 176 |
+
"br": 13,
|
| 177 |
+
"bs": 14,
|
| 178 |
+
"ca": 15,
|
| 179 |
+
"ceb": 16,
|
| 180 |
+
"ckb": 17,
|
| 181 |
+
"cs": 18,
|
| 182 |
+
"cy": 19,
|
| 183 |
+
"da": 20,
|
| 184 |
+
"de": 21,
|
| 185 |
+
"el": 22,
|
| 186 |
+
"en": 23,
|
| 187 |
+
"eo": 24,
|
| 188 |
+
"es": 25,
|
| 189 |
+
"et": 26,
|
| 190 |
+
"eu": 27,
|
| 191 |
+
"fa": 28,
|
| 192 |
+
"fi": 29,
|
| 193 |
+
"fr": 30,
|
| 194 |
+
"fy": 31,
|
| 195 |
+
"ga": 32,
|
| 196 |
+
"gan": 33,
|
| 197 |
+
"gl": 34,
|
| 198 |
+
"gu": 35,
|
| 199 |
+
"he": 36,
|
| 200 |
+
"hi": 37,
|
| 201 |
+
"hr": 38,
|
| 202 |
+
"hu": 39,
|
| 203 |
+
"hy": 40,
|
| 204 |
+
"ia": 41,
|
| 205 |
+
"id": 42,
|
| 206 |
+
"is": 43,
|
| 207 |
+
"it": 44,
|
| 208 |
+
"ja": 45,
|
| 209 |
+
"jv": 46,
|
| 210 |
+
"ka": 47,
|
| 211 |
+
"kk": 48,
|
| 212 |
+
"kn": 49,
|
| 213 |
+
"ko": 50,
|
| 214 |
+
"ku": 51,
|
| 215 |
+
"la": 52,
|
| 216 |
+
"lb": 53,
|
| 217 |
+
"lt": 54,
|
| 218 |
+
"lv": 55,
|
| 219 |
+
"mk": 56,
|
| 220 |
+
"ml": 57,
|
| 221 |
+
"mn": 58,
|
| 222 |
+
"mr": 59,
|
| 223 |
+
"ms": 60,
|
| 224 |
+
"my": 61,
|
| 225 |
+
"nds": 62,
|
| 226 |
+
"ne": 63,
|
| 227 |
+
"nl": 64,
|
| 228 |
+
"nn": 65,
|
| 229 |
+
"no": 66,
|
| 230 |
+
"oc": 67,
|
| 231 |
+
"pl": 68,
|
| 232 |
+
"pt": 69,
|
| 233 |
+
"ro": 70,
|
| 234 |
+
"ru": 71,
|
| 235 |
+
"scn": 72,
|
| 236 |
+
"sco": 73,
|
| 237 |
+
"sh": 74,
|
| 238 |
+
"si": 75,
|
| 239 |
+
"simple": 76,
|
| 240 |
+
"sk": 77,
|
| 241 |
+
"sl": 78,
|
| 242 |
+
"sq": 79,
|
| 243 |
+
"sr": 80,
|
| 244 |
+
"sv": 81,
|
| 245 |
+
"sw": 82,
|
| 246 |
+
"ta": 83,
|
| 247 |
+
"te": 84,
|
| 248 |
+
"th": 85,
|
| 249 |
+
"tl": 86,
|
| 250 |
+
"tr": 87,
|
| 251 |
+
"tt": 88,
|
| 252 |
+
"uk": 89,
|
| 253 |
+
"ur": 90,
|
| 254 |
+
"uz": 91,
|
| 255 |
+
"vi": 92,
|
| 256 |
+
"war": 93,
|
| 257 |
+
"wuu": 94,
|
| 258 |
+
"yi": 95,
|
| 259 |
+
"zh": 96,
|
| 260 |
+
"zh_classical": 97,
|
| 261 |
+
"zh_min_nan": 98,
|
| 262 |
+
"zh_yue": 99
|
| 263 |
+
},
|
| 264 |
+
"lang_id": 0,
|
| 265 |
+
"langs": [
|
| 266 |
+
"en",
|
| 267 |
+
"es",
|
| 268 |
+
"fr",
|
| 269 |
+
"de",
|
| 270 |
+
"zh",
|
| 271 |
+
"ru",
|
| 272 |
+
"pt",
|
| 273 |
+
"it",
|
| 274 |
+
"ar",
|
| 275 |
+
"ja",
|
| 276 |
+
"id",
|
| 277 |
+
"tr",
|
| 278 |
+
"nl",
|
| 279 |
+
"pl",
|
| 280 |
+
"simple",
|
| 281 |
+
"fa",
|
| 282 |
+
"vi",
|
| 283 |
+
"sv",
|
| 284 |
+
"ko",
|
| 285 |
+
"he",
|
| 286 |
+
"ro",
|
| 287 |
+
"no",
|
| 288 |
+
"hi",
|
| 289 |
+
"uk",
|
| 290 |
+
"cs",
|
| 291 |
+
"fi",
|
| 292 |
+
"hu",
|
| 293 |
+
"th",
|
| 294 |
+
"da",
|
| 295 |
+
"ca",
|
| 296 |
+
"el",
|
| 297 |
+
"bg",
|
| 298 |
+
"sr",
|
| 299 |
+
"ms",
|
| 300 |
+
"bn",
|
| 301 |
+
"hr",
|
| 302 |
+
"sl",
|
| 303 |
+
"zh_yue",
|
| 304 |
+
"az",
|
| 305 |
+
"sk",
|
| 306 |
+
"eo",
|
| 307 |
+
"ta",
|
| 308 |
+
"sh",
|
| 309 |
+
"lt",
|
| 310 |
+
"et",
|
| 311 |
+
"ml",
|
| 312 |
+
"la",
|
| 313 |
+
"bs",
|
| 314 |
+
"sq",
|
| 315 |
+
"arz",
|
| 316 |
+
"af",
|
| 317 |
+
"ka",
|
| 318 |
+
"mr",
|
| 319 |
+
"eu",
|
| 320 |
+
"tl",
|
| 321 |
+
"ang",
|
| 322 |
+
"gl",
|
| 323 |
+
"nn",
|
| 324 |
+
"ur",
|
| 325 |
+
"kk",
|
| 326 |
+
"be",
|
| 327 |
+
"hy",
|
| 328 |
+
"te",
|
| 329 |
+
"lv",
|
| 330 |
+
"mk",
|
| 331 |
+
"zh_classical",
|
| 332 |
+
"als",
|
| 333 |
+
"is",
|
| 334 |
+
"wuu",
|
| 335 |
+
"my",
|
| 336 |
+
"sco",
|
| 337 |
+
"mn",
|
| 338 |
+
"ceb",
|
| 339 |
+
"ast",
|
| 340 |
+
"cy",
|
| 341 |
+
"kn",
|
| 342 |
+
"br",
|
| 343 |
+
"an",
|
| 344 |
+
"gu",
|
| 345 |
+
"bar",
|
| 346 |
+
"uz",
|
| 347 |
+
"lb",
|
| 348 |
+
"ne",
|
| 349 |
+
"si",
|
| 350 |
+
"war",
|
| 351 |
+
"jv",
|
| 352 |
+
"ga",
|
| 353 |
+
"zh_min_nan",
|
| 354 |
+
"oc",
|
| 355 |
+
"ku",
|
| 356 |
+
"sw",
|
| 357 |
+
"nds",
|
| 358 |
+
"ckb",
|
| 359 |
+
"ia",
|
| 360 |
+
"yi",
|
| 361 |
+
"fy",
|
| 362 |
+
"scn",
|
| 363 |
+
"gan",
|
| 364 |
+
"tt",
|
| 365 |
+
"am"
|
| 366 |
+
],
|
| 367 |
+
"layer_norm_eps": 1e-12,
|
| 368 |
+
"lg_sampling_factor": 0.7,
|
| 369 |
+
"lgs": "en-es-fr-de-zh-ru-pt-it-ar-ja-id-tr-nl-pl-simple-fa-vi-sv-ko-he-ro-no-hi-uk-cs-fi-hu-th-da-ca-el-bg-sr-ms-bn-hr-sl-zh_yue-az-sk-eo-ta-sh-lt-et-ml-la-bs-sq-arz-af-ka-mr-eu-tl-ang-gl-nn-ur-kk-be-hy-te-lv-mk-zh_classical-als-is-wuu-my-sco-mn-ceb-ast-cy-kn-br-an-gu-bar-uz-lb-ne-si-war-jv-ga-zh_min_nan-oc-ku-sw-nds-ckb-ia-yi-fy-scn-gan-tt-am",
|
| 370 |
+
"local_rank": 0,
|
| 371 |
+
"mask_index": 5,
|
| 372 |
+
"mask_token_id": 0,
|
| 373 |
+
"master_addr": "learnfair0332",
|
| 374 |
+
"master_port": 11363,
|
| 375 |
+
"max_batch_size": 0,
|
| 376 |
+
"max_epoch": 100000,
|
| 377 |
+
"max_len": 200,
|
| 378 |
+
"max_position_embeddings": 512,
|
| 379 |
+
"max_vocab": 200000,
|
| 380 |
+
"min_count": 0,
|
| 381 |
+
"mlm_steps": [
|
| 382 |
+
[
|
| 383 |
+
"en",
|
| 384 |
+
null
|
| 385 |
+
],
|
| 386 |
+
[
|
| 387 |
+
"es",
|
| 388 |
+
null
|
| 389 |
+
],
|
| 390 |
+
[
|
| 391 |
+
"fr",
|
| 392 |
+
null
|
| 393 |
+
],
|
| 394 |
+
[
|
| 395 |
+
"de",
|
| 396 |
+
null
|
| 397 |
+
],
|
| 398 |
+
[
|
| 399 |
+
"zh",
|
| 400 |
+
null
|
| 401 |
+
],
|
| 402 |
+
[
|
| 403 |
+
"ru",
|
| 404 |
+
null
|
| 405 |
+
],
|
| 406 |
+
[
|
| 407 |
+
"pt",
|
| 408 |
+
null
|
| 409 |
+
],
|
| 410 |
+
[
|
| 411 |
+
"it",
|
| 412 |
+
null
|
| 413 |
+
],
|
| 414 |
+
[
|
| 415 |
+
"ar",
|
| 416 |
+
null
|
| 417 |
+
],
|
| 418 |
+
[
|
| 419 |
+
"ja",
|
| 420 |
+
null
|
| 421 |
+
],
|
| 422 |
+
[
|
| 423 |
+
"id",
|
| 424 |
+
null
|
| 425 |
+
],
|
| 426 |
+
[
|
| 427 |
+
"tr",
|
| 428 |
+
null
|
| 429 |
+
],
|
| 430 |
+
[
|
| 431 |
+
"nl",
|
| 432 |
+
null
|
| 433 |
+
],
|
| 434 |
+
[
|
| 435 |
+
"pl",
|
| 436 |
+
null
|
| 437 |
+
],
|
| 438 |
+
[
|
| 439 |
+
"simple",
|
| 440 |
+
null
|
| 441 |
+
],
|
| 442 |
+
[
|
| 443 |
+
"fa",
|
| 444 |
+
null
|
| 445 |
+
],
|
| 446 |
+
[
|
| 447 |
+
"vi",
|
| 448 |
+
null
|
| 449 |
+
],
|
| 450 |
+
[
|
| 451 |
+
"sv",
|
| 452 |
+
null
|
| 453 |
+
],
|
| 454 |
+
[
|
| 455 |
+
"ko",
|
| 456 |
+
null
|
| 457 |
+
],
|
| 458 |
+
[
|
| 459 |
+
"he",
|
| 460 |
+
null
|
| 461 |
+
],
|
| 462 |
+
[
|
| 463 |
+
"ro",
|
| 464 |
+
null
|
| 465 |
+
],
|
| 466 |
+
[
|
| 467 |
+
"no",
|
| 468 |
+
null
|
| 469 |
+
],
|
| 470 |
+
[
|
| 471 |
+
"hi",
|
| 472 |
+
null
|
| 473 |
+
],
|
| 474 |
+
[
|
| 475 |
+
"uk",
|
| 476 |
+
null
|
| 477 |
+
],
|
| 478 |
+
[
|
| 479 |
+
"cs",
|
| 480 |
+
null
|
| 481 |
+
],
|
| 482 |
+
[
|
| 483 |
+
"fi",
|
| 484 |
+
null
|
| 485 |
+
],
|
| 486 |
+
[
|
| 487 |
+
"hu",
|
| 488 |
+
null
|
| 489 |
+
],
|
| 490 |
+
[
|
| 491 |
+
"th",
|
| 492 |
+
null
|
| 493 |
+
],
|
| 494 |
+
[
|
| 495 |
+
"da",
|
| 496 |
+
null
|
| 497 |
+
],
|
| 498 |
+
[
|
| 499 |
+
"ca",
|
| 500 |
+
null
|
| 501 |
+
],
|
| 502 |
+
[
|
| 503 |
+
"el",
|
| 504 |
+
null
|
| 505 |
+
],
|
| 506 |
+
[
|
| 507 |
+
"bg",
|
| 508 |
+
null
|
| 509 |
+
],
|
| 510 |
+
[
|
| 511 |
+
"sr",
|
| 512 |
+
null
|
| 513 |
+
],
|
| 514 |
+
[
|
| 515 |
+
"ms",
|
| 516 |
+
null
|
| 517 |
+
],
|
| 518 |
+
[
|
| 519 |
+
"bn",
|
| 520 |
+
null
|
| 521 |
+
],
|
| 522 |
+
[
|
| 523 |
+
"hr",
|
| 524 |
+
null
|
| 525 |
+
],
|
| 526 |
+
[
|
| 527 |
+
"sl",
|
| 528 |
+
null
|
| 529 |
+
],
|
| 530 |
+
[
|
| 531 |
+
"zh_yue",
|
| 532 |
+
null
|
| 533 |
+
],
|
| 534 |
+
[
|
| 535 |
+
"az",
|
| 536 |
+
null
|
| 537 |
+
],
|
| 538 |
+
[
|
| 539 |
+
"sk",
|
| 540 |
+
null
|
| 541 |
+
],
|
| 542 |
+
[
|
| 543 |
+
"eo",
|
| 544 |
+
null
|
| 545 |
+
],
|
| 546 |
+
[
|
| 547 |
+
"ta",
|
| 548 |
+
null
|
| 549 |
+
],
|
| 550 |
+
[
|
| 551 |
+
"sh",
|
| 552 |
+
null
|
| 553 |
+
],
|
| 554 |
+
[
|
| 555 |
+
"lt",
|
| 556 |
+
null
|
| 557 |
+
],
|
| 558 |
+
[
|
| 559 |
+
"et",
|
| 560 |
+
null
|
| 561 |
+
],
|
| 562 |
+
[
|
| 563 |
+
"ml",
|
| 564 |
+
null
|
| 565 |
+
],
|
| 566 |
+
[
|
| 567 |
+
"la",
|
| 568 |
+
null
|
| 569 |
+
],
|
| 570 |
+
[
|
| 571 |
+
"bs",
|
| 572 |
+
null
|
| 573 |
+
],
|
| 574 |
+
[
|
| 575 |
+
"sq",
|
| 576 |
+
null
|
| 577 |
+
],
|
| 578 |
+
[
|
| 579 |
+
"arz",
|
| 580 |
+
null
|
| 581 |
+
],
|
| 582 |
+
[
|
| 583 |
+
"af",
|
| 584 |
+
null
|
| 585 |
+
],
|
| 586 |
+
[
|
| 587 |
+
"ka",
|
| 588 |
+
null
|
| 589 |
+
],
|
| 590 |
+
[
|
| 591 |
+
"mr",
|
| 592 |
+
null
|
| 593 |
+
],
|
| 594 |
+
[
|
| 595 |
+
"eu",
|
| 596 |
+
null
|
| 597 |
+
],
|
| 598 |
+
[
|
| 599 |
+
"tl",
|
| 600 |
+
null
|
| 601 |
+
],
|
| 602 |
+
[
|
| 603 |
+
"ang",
|
| 604 |
+
null
|
| 605 |
+
],
|
| 606 |
+
[
|
| 607 |
+
"gl",
|
| 608 |
+
null
|
| 609 |
+
],
|
| 610 |
+
[
|
| 611 |
+
"nn",
|
| 612 |
+
null
|
| 613 |
+
],
|
| 614 |
+
[
|
| 615 |
+
"ur",
|
| 616 |
+
null
|
| 617 |
+
],
|
| 618 |
+
[
|
| 619 |
+
"kk",
|
| 620 |
+
null
|
| 621 |
+
],
|
| 622 |
+
[
|
| 623 |
+
"be",
|
| 624 |
+
null
|
| 625 |
+
],
|
| 626 |
+
[
|
| 627 |
+
"hy",
|
| 628 |
+
null
|
| 629 |
+
],
|
| 630 |
+
[
|
| 631 |
+
"te",
|
| 632 |
+
null
|
| 633 |
+
],
|
| 634 |
+
[
|
| 635 |
+
"lv",
|
| 636 |
+
null
|
| 637 |
+
],
|
| 638 |
+
[
|
| 639 |
+
"mk",
|
| 640 |
+
null
|
| 641 |
+
],
|
| 642 |
+
[
|
| 643 |
+
"zh_classical",
|
| 644 |
+
null
|
| 645 |
+
],
|
| 646 |
+
[
|
| 647 |
+
"als",
|
| 648 |
+
null
|
| 649 |
+
],
|
| 650 |
+
[
|
| 651 |
+
"is",
|
| 652 |
+
null
|
| 653 |
+
],
|
| 654 |
+
[
|
| 655 |
+
"wuu",
|
| 656 |
+
null
|
| 657 |
+
],
|
| 658 |
+
[
|
| 659 |
+
"my",
|
| 660 |
+
null
|
| 661 |
+
],
|
| 662 |
+
[
|
| 663 |
+
"sco",
|
| 664 |
+
null
|
| 665 |
+
],
|
| 666 |
+
[
|
| 667 |
+
"mn",
|
| 668 |
+
null
|
| 669 |
+
],
|
| 670 |
+
[
|
| 671 |
+
"ceb",
|
| 672 |
+
null
|
| 673 |
+
],
|
| 674 |
+
[
|
| 675 |
+
"ast",
|
| 676 |
+
null
|
| 677 |
+
],
|
| 678 |
+
[
|
| 679 |
+
"cy",
|
| 680 |
+
null
|
| 681 |
+
],
|
| 682 |
+
[
|
| 683 |
+
"kn",
|
| 684 |
+
null
|
| 685 |
+
],
|
| 686 |
+
[
|
| 687 |
+
"br",
|
| 688 |
+
null
|
| 689 |
+
],
|
| 690 |
+
[
|
| 691 |
+
"an",
|
| 692 |
+
null
|
| 693 |
+
],
|
| 694 |
+
[
|
| 695 |
+
"gu",
|
| 696 |
+
null
|
| 697 |
+
],
|
| 698 |
+
[
|
| 699 |
+
"bar",
|
| 700 |
+
null
|
| 701 |
+
],
|
| 702 |
+
[
|
| 703 |
+
"uz",
|
| 704 |
+
null
|
| 705 |
+
],
|
| 706 |
+
[
|
| 707 |
+
"lb",
|
| 708 |
+
null
|
| 709 |
+
],
|
| 710 |
+
[
|
| 711 |
+
"ne",
|
| 712 |
+
null
|
| 713 |
+
],
|
| 714 |
+
[
|
| 715 |
+
"si",
|
| 716 |
+
null
|
| 717 |
+
],
|
| 718 |
+
[
|
| 719 |
+
"war",
|
| 720 |
+
null
|
| 721 |
+
],
|
| 722 |
+
[
|
| 723 |
+
"jv",
|
| 724 |
+
null
|
| 725 |
+
],
|
| 726 |
+
[
|
| 727 |
+
"ga",
|
| 728 |
+
null
|
| 729 |
+
],
|
| 730 |
+
[
|
| 731 |
+
"zh_min_nan",
|
| 732 |
+
null
|
| 733 |
+
],
|
| 734 |
+
[
|
| 735 |
+
"oc",
|
| 736 |
+
null
|
| 737 |
+
],
|
| 738 |
+
[
|
| 739 |
+
"ku",
|
| 740 |
+
null
|
| 741 |
+
],
|
| 742 |
+
[
|
| 743 |
+
"sw",
|
| 744 |
+
null
|
| 745 |
+
],
|
| 746 |
+
[
|
| 747 |
+
"nds",
|
| 748 |
+
null
|
| 749 |
+
],
|
| 750 |
+
[
|
| 751 |
+
"ckb",
|
| 752 |
+
null
|
| 753 |
+
],
|
| 754 |
+
[
|
| 755 |
+
"ia",
|
| 756 |
+
null
|
| 757 |
+
],
|
| 758 |
+
[
|
| 759 |
+
"yi",
|
| 760 |
+
null
|
| 761 |
+
],
|
| 762 |
+
[
|
| 763 |
+
"fy",
|
| 764 |
+
null
|
| 765 |
+
],
|
| 766 |
+
[
|
| 767 |
+
"scn",
|
| 768 |
+
null
|
| 769 |
+
],
|
| 770 |
+
[
|
| 771 |
+
"gan",
|
| 772 |
+
null
|
| 773 |
+
],
|
| 774 |
+
[
|
| 775 |
+
"tt",
|
| 776 |
+
null
|
| 777 |
+
],
|
| 778 |
+
[
|
| 779 |
+
"am",
|
| 780 |
+
null
|
| 781 |
+
]
|
| 782 |
+
],
|
| 783 |
+
"model_type": "xlm",
|
| 784 |
+
"mono_dataset": {
|
| 785 |
+
"af": {
|
| 786 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.af.pth",
|
| 787 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.af.pth",
|
| 788 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.af.pth"
|
| 789 |
+
},
|
| 790 |
+
"als": {
|
| 791 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.als.pth",
|
| 792 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.als.pth",
|
| 793 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.als.pth"
|
| 794 |
+
},
|
| 795 |
+
"am": {
|
| 796 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.am.pth",
|
| 797 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.am.pth",
|
| 798 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.am.pth"
|
| 799 |
+
},
|
| 800 |
+
"an": {
|
| 801 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.an.pth",
|
| 802 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.an.pth",
|
| 803 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.an.pth"
|
| 804 |
+
},
|
| 805 |
+
"ang": {
|
| 806 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ang.pth",
|
| 807 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ang.pth",
|
| 808 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ang.pth"
|
| 809 |
+
},
|
| 810 |
+
"ar": {
|
| 811 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ar.pth",
|
| 812 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ar.pth",
|
| 813 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ar.pth"
|
| 814 |
+
},
|
| 815 |
+
"arz": {
|
| 816 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.arz.pth",
|
| 817 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.arz.pth",
|
| 818 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.arz.pth"
|
| 819 |
+
},
|
| 820 |
+
"ast": {
|
| 821 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ast.pth",
|
| 822 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ast.pth",
|
| 823 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ast.pth"
|
| 824 |
+
},
|
| 825 |
+
"az": {
|
| 826 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.az.pth",
|
| 827 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.az.pth",
|
| 828 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.az.pth"
|
| 829 |
+
},
|
| 830 |
+
"bar": {
|
| 831 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.bar.pth",
|
| 832 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.bar.pth",
|
| 833 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.bar.pth"
|
| 834 |
+
},
|
| 835 |
+
"be": {
|
| 836 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.be.pth",
|
| 837 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.be.pth",
|
| 838 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.be.pth"
|
| 839 |
+
},
|
| 840 |
+
"bg": {
|
| 841 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.bg.pth",
|
| 842 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.bg.pth",
|
| 843 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.bg.pth"
|
| 844 |
+
},
|
| 845 |
+
"bn": {
|
| 846 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.bn.pth",
|
| 847 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.bn.pth",
|
| 848 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.bn.pth"
|
| 849 |
+
},
|
| 850 |
+
"br": {
|
| 851 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.br.pth",
|
| 852 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.br.pth",
|
| 853 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.br.pth"
|
| 854 |
+
},
|
| 855 |
+
"bs": {
|
| 856 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.bs.pth",
|
| 857 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.bs.pth",
|
| 858 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.bs.pth"
|
| 859 |
+
},
|
| 860 |
+
"ca": {
|
| 861 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ca.pth",
|
| 862 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ca.pth",
|
| 863 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ca.pth"
|
| 864 |
+
},
|
| 865 |
+
"ceb": {
|
| 866 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ceb.pth",
|
| 867 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ceb.pth",
|
| 868 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ceb.pth"
|
| 869 |
+
},
|
| 870 |
+
"ckb": {
|
| 871 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ckb.pth",
|
| 872 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ckb.pth",
|
| 873 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ckb.pth"
|
| 874 |
+
},
|
| 875 |
+
"cs": {
|
| 876 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.cs.pth",
|
| 877 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.cs.pth",
|
| 878 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.cs.pth"
|
| 879 |
+
},
|
| 880 |
+
"cy": {
|
| 881 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.cy.pth",
|
| 882 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.cy.pth",
|
| 883 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.cy.pth"
|
| 884 |
+
},
|
| 885 |
+
"da": {
|
| 886 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.da.pth",
|
| 887 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.da.pth",
|
| 888 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.da.pth"
|
| 889 |
+
},
|
| 890 |
+
"de": {
|
| 891 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.de.pth",
|
| 892 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.de.pth",
|
| 893 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.de.pth"
|
| 894 |
+
},
|
| 895 |
+
"el": {
|
| 896 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.el.pth",
|
| 897 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.el.pth",
|
| 898 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.el.pth"
|
| 899 |
+
},
|
| 900 |
+
"en": {
|
| 901 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.en.pth",
|
| 902 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.en.pth",
|
| 903 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.en.pth"
|
| 904 |
+
},
|
| 905 |
+
"eo": {
|
| 906 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.eo.pth",
|
| 907 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.eo.pth",
|
| 908 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.eo.pth"
|
| 909 |
+
},
|
| 910 |
+
"es": {
|
| 911 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.es.pth",
|
| 912 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.es.pth",
|
| 913 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.es.pth"
|
| 914 |
+
},
|
| 915 |
+
"et": {
|
| 916 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.et.pth",
|
| 917 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.et.pth",
|
| 918 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.et.pth"
|
| 919 |
+
},
|
| 920 |
+
"eu": {
|
| 921 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.eu.pth",
|
| 922 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.eu.pth",
|
| 923 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.eu.pth"
|
| 924 |
+
},
|
| 925 |
+
"fa": {
|
| 926 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.fa.pth",
|
| 927 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.fa.pth",
|
| 928 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.fa.pth"
|
| 929 |
+
},
|
| 930 |
+
"fi": {
|
| 931 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.fi.pth",
|
| 932 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.fi.pth",
|
| 933 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.fi.pth"
|
| 934 |
+
},
|
| 935 |
+
"fr": {
|
| 936 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.fr.pth",
|
| 937 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.fr.pth",
|
| 938 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.fr.pth"
|
| 939 |
+
},
|
| 940 |
+
"fy": {
|
| 941 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.fy.pth",
|
| 942 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.fy.pth",
|
| 943 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.fy.pth"
|
| 944 |
+
},
|
| 945 |
+
"ga": {
|
| 946 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ga.pth",
|
| 947 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ga.pth",
|
| 948 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ga.pth"
|
| 949 |
+
},
|
| 950 |
+
"gan": {
|
| 951 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.gan.pth",
|
| 952 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.gan.pth",
|
| 953 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.gan.pth"
|
| 954 |
+
},
|
| 955 |
+
"gl": {
|
| 956 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.gl.pth",
|
| 957 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.gl.pth",
|
| 958 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.gl.pth"
|
| 959 |
+
},
|
| 960 |
+
"gu": {
|
| 961 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.gu.pth",
|
| 962 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.gu.pth",
|
| 963 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.gu.pth"
|
| 964 |
+
},
|
| 965 |
+
"he": {
|
| 966 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.he.pth",
|
| 967 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.he.pth",
|
| 968 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.he.pth"
|
| 969 |
+
},
|
| 970 |
+
"hi": {
|
| 971 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.hi.pth",
|
| 972 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.hi.pth",
|
| 973 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.hi.pth"
|
| 974 |
+
},
|
| 975 |
+
"hr": {
|
| 976 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.hr.pth",
|
| 977 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.hr.pth",
|
| 978 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.hr.pth"
|
| 979 |
+
},
|
| 980 |
+
"hu": {
|
| 981 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.hu.pth",
|
| 982 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.hu.pth",
|
| 983 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.hu.pth"
|
| 984 |
+
},
|
| 985 |
+
"hy": {
|
| 986 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.hy.pth",
|
| 987 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.hy.pth",
|
| 988 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.hy.pth"
|
| 989 |
+
},
|
| 990 |
+
"ia": {
|
| 991 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ia.pth",
|
| 992 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ia.pth",
|
| 993 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ia.pth"
|
| 994 |
+
},
|
| 995 |
+
"id": {
|
| 996 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.id.pth",
|
| 997 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.id.pth",
|
| 998 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.id.pth"
|
| 999 |
+
},
|
| 1000 |
+
"is": {
|
| 1001 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.is.pth",
|
| 1002 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.is.pth",
|
| 1003 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.is.pth"
|
| 1004 |
+
},
|
| 1005 |
+
"it": {
|
| 1006 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.it.pth",
|
| 1007 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.it.pth",
|
| 1008 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.it.pth"
|
| 1009 |
+
},
|
| 1010 |
+
"ja": {
|
| 1011 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ja.pth",
|
| 1012 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ja.pth",
|
| 1013 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ja.pth"
|
| 1014 |
+
},
|
| 1015 |
+
"jv": {
|
| 1016 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.jv.pth",
|
| 1017 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.jv.pth",
|
| 1018 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.jv.pth"
|
| 1019 |
+
},
|
| 1020 |
+
"ka": {
|
| 1021 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ka.pth",
|
| 1022 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ka.pth",
|
| 1023 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ka.pth"
|
| 1024 |
+
},
|
| 1025 |
+
"kk": {
|
| 1026 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.kk.pth",
|
| 1027 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.kk.pth",
|
| 1028 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.kk.pth"
|
| 1029 |
+
},
|
| 1030 |
+
"kn": {
|
| 1031 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.kn.pth",
|
| 1032 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.kn.pth",
|
| 1033 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.kn.pth"
|
| 1034 |
+
},
|
| 1035 |
+
"ko": {
|
| 1036 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ko.pth",
|
| 1037 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ko.pth",
|
| 1038 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ko.pth"
|
| 1039 |
+
},
|
| 1040 |
+
"ku": {
|
| 1041 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ku.pth",
|
| 1042 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ku.pth",
|
| 1043 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ku.pth"
|
| 1044 |
+
},
|
| 1045 |
+
"la": {
|
| 1046 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.la.pth",
|
| 1047 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.la.pth",
|
| 1048 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.la.pth"
|
| 1049 |
+
},
|
| 1050 |
+
"lb": {
|
| 1051 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.lb.pth",
|
| 1052 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.lb.pth",
|
| 1053 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.lb.pth"
|
| 1054 |
+
},
|
| 1055 |
+
"lt": {
|
| 1056 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.lt.pth",
|
| 1057 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.lt.pth",
|
| 1058 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.lt.pth"
|
| 1059 |
+
},
|
| 1060 |
+
"lv": {
|
| 1061 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.lv.pth",
|
| 1062 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.lv.pth",
|
| 1063 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.lv.pth"
|
| 1064 |
+
},
|
| 1065 |
+
"mk": {
|
| 1066 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.mk.pth",
|
| 1067 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.mk.pth",
|
| 1068 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.mk.pth"
|
| 1069 |
+
},
|
| 1070 |
+
"ml": {
|
| 1071 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ml.pth",
|
| 1072 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ml.pth",
|
| 1073 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ml.pth"
|
| 1074 |
+
},
|
| 1075 |
+
"mn": {
|
| 1076 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.mn.pth",
|
| 1077 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.mn.pth",
|
| 1078 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.mn.pth"
|
| 1079 |
+
},
|
| 1080 |
+
"mr": {
|
| 1081 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.mr.pth",
|
| 1082 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.mr.pth",
|
| 1083 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.mr.pth"
|
| 1084 |
+
},
|
| 1085 |
+
"ms": {
|
| 1086 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ms.pth",
|
| 1087 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ms.pth",
|
| 1088 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ms.pth"
|
| 1089 |
+
},
|
| 1090 |
+
"my": {
|
| 1091 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.my.pth",
|
| 1092 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.my.pth",
|
| 1093 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.my.pth"
|
| 1094 |
+
},
|
| 1095 |
+
"nds": {
|
| 1096 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.nds.pth",
|
| 1097 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.nds.pth",
|
| 1098 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.nds.pth"
|
| 1099 |
+
},
|
| 1100 |
+
"ne": {
|
| 1101 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ne.pth",
|
| 1102 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ne.pth",
|
| 1103 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ne.pth"
|
| 1104 |
+
},
|
| 1105 |
+
"nl": {
|
| 1106 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.nl.pth",
|
| 1107 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.nl.pth",
|
| 1108 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.nl.pth"
|
| 1109 |
+
},
|
| 1110 |
+
"nn": {
|
| 1111 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.nn.pth",
|
| 1112 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.nn.pth",
|
| 1113 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.nn.pth"
|
| 1114 |
+
},
|
| 1115 |
+
"no": {
|
| 1116 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.no.pth",
|
| 1117 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.no.pth",
|
| 1118 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.no.pth"
|
| 1119 |
+
},
|
| 1120 |
+
"oc": {
|
| 1121 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.oc.pth",
|
| 1122 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.oc.pth",
|
| 1123 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.oc.pth"
|
| 1124 |
+
},
|
| 1125 |
+
"pl": {
|
| 1126 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.pl.pth",
|
| 1127 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.pl.pth",
|
| 1128 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.pl.pth"
|
| 1129 |
+
},
|
| 1130 |
+
"pt": {
|
| 1131 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.pt.pth",
|
| 1132 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.pt.pth",
|
| 1133 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.pt.pth"
|
| 1134 |
+
},
|
| 1135 |
+
"ro": {
|
| 1136 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ro.pth",
|
| 1137 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ro.pth",
|
| 1138 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ro.pth"
|
| 1139 |
+
},
|
| 1140 |
+
"ru": {
|
| 1141 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ru.pth",
|
| 1142 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ru.pth",
|
| 1143 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ru.pth"
|
| 1144 |
+
},
|
| 1145 |
+
"scn": {
|
| 1146 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.scn.pth",
|
| 1147 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.scn.pth",
|
| 1148 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.scn.pth"
|
| 1149 |
+
},
|
| 1150 |
+
"sco": {
|
| 1151 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sco.pth",
|
| 1152 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sco.pth",
|
| 1153 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sco.pth"
|
| 1154 |
+
},
|
| 1155 |
+
"sh": {
|
| 1156 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sh.pth",
|
| 1157 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sh.pth",
|
| 1158 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sh.pth"
|
| 1159 |
+
},
|
| 1160 |
+
"si": {
|
| 1161 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.si.pth",
|
| 1162 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.si.pth",
|
| 1163 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.si.pth"
|
| 1164 |
+
},
|
| 1165 |
+
"simple": {
|
| 1166 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.simple.pth",
|
| 1167 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.simple.pth",
|
| 1168 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.simple.pth"
|
| 1169 |
+
},
|
| 1170 |
+
"sk": {
|
| 1171 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sk.pth",
|
| 1172 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sk.pth",
|
| 1173 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sk.pth"
|
| 1174 |
+
},
|
| 1175 |
+
"sl": {
|
| 1176 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sl.pth",
|
| 1177 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sl.pth",
|
| 1178 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sl.pth"
|
| 1179 |
+
},
|
| 1180 |
+
"sq": {
|
| 1181 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sq.pth",
|
| 1182 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sq.pth",
|
| 1183 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sq.pth"
|
| 1184 |
+
},
|
| 1185 |
+
"sr": {
|
| 1186 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sr.pth",
|
| 1187 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sr.pth",
|
| 1188 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sr.pth"
|
| 1189 |
+
},
|
| 1190 |
+
"sv": {
|
| 1191 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sv.pth",
|
| 1192 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sv.pth",
|
| 1193 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sv.pth"
|
| 1194 |
+
},
|
| 1195 |
+
"sw": {
|
| 1196 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.sw.pth",
|
| 1197 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.sw.pth",
|
| 1198 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.sw.pth"
|
| 1199 |
+
},
|
| 1200 |
+
"ta": {
|
| 1201 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ta.pth",
|
| 1202 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ta.pth",
|
| 1203 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ta.pth"
|
| 1204 |
+
},
|
| 1205 |
+
"te": {
|
| 1206 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.te.pth",
|
| 1207 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.te.pth",
|
| 1208 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.te.pth"
|
| 1209 |
+
},
|
| 1210 |
+
"th": {
|
| 1211 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.th.pth",
|
| 1212 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.th.pth",
|
| 1213 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.th.pth"
|
| 1214 |
+
},
|
| 1215 |
+
"tl": {
|
| 1216 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.tl.pth",
|
| 1217 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.tl.pth",
|
| 1218 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.tl.pth"
|
| 1219 |
+
},
|
| 1220 |
+
"tr": {
|
| 1221 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.tr.pth",
|
| 1222 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.tr.pth",
|
| 1223 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.tr.pth"
|
| 1224 |
+
},
|
| 1225 |
+
"tt": {
|
| 1226 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.tt.pth",
|
| 1227 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.tt.pth",
|
| 1228 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.tt.pth"
|
| 1229 |
+
},
|
| 1230 |
+
"uk": {
|
| 1231 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.uk.pth",
|
| 1232 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.uk.pth",
|
| 1233 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.uk.pth"
|
| 1234 |
+
},
|
| 1235 |
+
"ur": {
|
| 1236 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.ur.pth",
|
| 1237 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.ur.pth",
|
| 1238 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.ur.pth"
|
| 1239 |
+
},
|
| 1240 |
+
"uz": {
|
| 1241 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.uz.pth",
|
| 1242 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.uz.pth",
|
| 1243 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.uz.pth"
|
| 1244 |
+
},
|
| 1245 |
+
"vi": {
|
| 1246 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.vi.pth",
|
| 1247 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.vi.pth",
|
| 1248 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.vi.pth"
|
| 1249 |
+
},
|
| 1250 |
+
"war": {
|
| 1251 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.war.pth",
|
| 1252 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.war.pth",
|
| 1253 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.war.pth"
|
| 1254 |
+
},
|
| 1255 |
+
"wuu": {
|
| 1256 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.wuu.pth",
|
| 1257 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.wuu.pth",
|
| 1258 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.wuu.pth"
|
| 1259 |
+
},
|
| 1260 |
+
"yi": {
|
| 1261 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.yi.pth",
|
| 1262 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.yi.pth",
|
| 1263 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.yi.pth"
|
| 1264 |
+
},
|
| 1265 |
+
"zh": {
|
| 1266 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.zh.pth",
|
| 1267 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.zh.pth",
|
| 1268 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.zh.pth"
|
| 1269 |
+
},
|
| 1270 |
+
"zh_classical": {
|
| 1271 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.zh_classical.pth",
|
| 1272 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.zh_classical.pth",
|
| 1273 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.zh_classical.pth"
|
| 1274 |
+
},
|
| 1275 |
+
"zh_min_nan": {
|
| 1276 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.zh_min_nan.pth",
|
| 1277 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.zh_min_nan.pth",
|
| 1278 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.zh_min_nan.pth"
|
| 1279 |
+
},
|
| 1280 |
+
"zh_yue": {
|
| 1281 |
+
"test": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/test.zh_yue.pth",
|
| 1282 |
+
"train": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/train.zh_yue.pth",
|
| 1283 |
+
"valid": "/private/home/aconneau/projects/XLM/data/wiki/100/175k/valid.zh_yue.pth"
|
| 1284 |
+
}
|
| 1285 |
+
},
|
| 1286 |
+
"mono_list": [
|
| 1287 |
+
"en",
|
| 1288 |
+
"es",
|
| 1289 |
+
"fr",
|
| 1290 |
+
"de",
|
| 1291 |
+
"zh",
|
| 1292 |
+
"ru",
|
| 1293 |
+
"pt",
|
| 1294 |
+
"it",
|
| 1295 |
+
"ar",
|
| 1296 |
+
"ja",
|
| 1297 |
+
"id",
|
| 1298 |
+
"tr",
|
| 1299 |
+
"nl",
|
| 1300 |
+
"pl",
|
| 1301 |
+
"simple",
|
| 1302 |
+
"fa",
|
| 1303 |
+
"vi",
|
| 1304 |
+
"sv",
|
| 1305 |
+
"ko",
|
| 1306 |
+
"he",
|
| 1307 |
+
"ro",
|
| 1308 |
+
"no",
|
| 1309 |
+
"hi",
|
| 1310 |
+
"uk",
|
| 1311 |
+
"cs",
|
| 1312 |
+
"fi",
|
| 1313 |
+
"hu",
|
| 1314 |
+
"th",
|
| 1315 |
+
"da",
|
| 1316 |
+
"ca",
|
| 1317 |
+
"el",
|
| 1318 |
+
"bg",
|
| 1319 |
+
"sr",
|
| 1320 |
+
"ms",
|
| 1321 |
+
"bn",
|
| 1322 |
+
"hr",
|
| 1323 |
+
"sl",
|
| 1324 |
+
"zh_yue",
|
| 1325 |
+
"az",
|
| 1326 |
+
"sk",
|
| 1327 |
+
"eo",
|
| 1328 |
+
"ta",
|
| 1329 |
+
"sh",
|
| 1330 |
+
"lt",
|
| 1331 |
+
"et",
|
| 1332 |
+
"ml",
|
| 1333 |
+
"la",
|
| 1334 |
+
"bs",
|
| 1335 |
+
"sq",
|
| 1336 |
+
"arz",
|
| 1337 |
+
"af",
|
| 1338 |
+
"ka",
|
| 1339 |
+
"mr",
|
| 1340 |
+
"eu",
|
| 1341 |
+
"tl",
|
| 1342 |
+
"ang",
|
| 1343 |
+
"gl",
|
| 1344 |
+
"nn",
|
| 1345 |
+
"ur",
|
| 1346 |
+
"kk",
|
| 1347 |
+
"be",
|
| 1348 |
+
"hy",
|
| 1349 |
+
"te",
|
| 1350 |
+
"lv",
|
| 1351 |
+
"mk",
|
| 1352 |
+
"zh_classical",
|
| 1353 |
+
"als",
|
| 1354 |
+
"is",
|
| 1355 |
+
"wuu",
|
| 1356 |
+
"my",
|
| 1357 |
+
"sco",
|
| 1358 |
+
"mn",
|
| 1359 |
+
"ceb",
|
| 1360 |
+
"ast",
|
| 1361 |
+
"cy",
|
| 1362 |
+
"kn",
|
| 1363 |
+
"br",
|
| 1364 |
+
"an",
|
| 1365 |
+
"gu",
|
| 1366 |
+
"bar",
|
| 1367 |
+
"uz",
|
| 1368 |
+
"lb",
|
| 1369 |
+
"ne",
|
| 1370 |
+
"si",
|
| 1371 |
+
"war",
|
| 1372 |
+
"jv",
|
| 1373 |
+
"ga",
|
| 1374 |
+
"zh_min_nan",
|
| 1375 |
+
"oc",
|
| 1376 |
+
"ku",
|
| 1377 |
+
"sw",
|
| 1378 |
+
"nds",
|
| 1379 |
+
"ckb",
|
| 1380 |
+
"ia",
|
| 1381 |
+
"yi",
|
| 1382 |
+
"fy",
|
| 1383 |
+
"scn",
|
| 1384 |
+
"gan",
|
| 1385 |
+
"tt",
|
| 1386 |
+
"am"
|
| 1387 |
+
],
|
| 1388 |
+
"mt_steps": [],
|
| 1389 |
+
"multi_gpu": true,
|
| 1390 |
+
"multi_node": true,
|
| 1391 |
+
"n_gpu_per_node": 8,
|
| 1392 |
+
"n_heads": 16,
|
| 1393 |
+
"n_langs": 100,
|
| 1394 |
+
"n_layers": 16,
|
| 1395 |
+
"n_nodes": 4,
|
| 1396 |
+
"node_id": 0,
|
| 1397 |
+
"optimizer": "adam_inverse_sqrt,lr=0.00005,warmup_updates=30000,beta1=0.9,beta2=0.999,weight_decay=0.01,eps=0.000001",
|
| 1398 |
+
"pad_index": 2,
|
| 1399 |
+
"pad_token_id": 2,
|
| 1400 |
+
"para_dataset": {},
|
| 1401 |
+
"para_list": [],
|
| 1402 |
+
"pc_steps": [],
|
| 1403 |
+
"ref_paths": {},
|
| 1404 |
+
"reload_checkpoint": "",
|
| 1405 |
+
"reload_emb": "",
|
| 1406 |
+
"reload_model": "/checkpoint/aconneau/dumped/xlm_17_100_240_big_model_upper.2/14884511/best-valid_zh_mlm_ppl.pth",
|
| 1407 |
+
"sample_alpha": 0.5,
|
| 1408 |
+
"save_periodic": 0,
|
| 1409 |
+
"share_inout_emb": true,
|
| 1410 |
+
"sinusoidal_embeddings": false,
|
| 1411 |
+
"split_data": true,
|
| 1412 |
+
"start_n_top": 5,
|
| 1413 |
+
"stopping_criterion": "_valid_zh_mlm_ppl,25",
|
| 1414 |
+
"summary_activation": null,
|
| 1415 |
+
"summary_first_dropout": 0.1,
|
| 1416 |
+
"summary_proj_to_labels": true,
|
| 1417 |
+
"summary_type": "first",
|
| 1418 |
+
"summary_use_proj": true,
|
| 1419 |
+
"tokens_per_batch": -1,
|
| 1420 |
+
"transformers_version": "4.34.0.dev0",
|
| 1421 |
+
"unk_index": 3,
|
| 1422 |
+
"use_lang_emb": false,
|
| 1423 |
+
"use_memory": false,
|
| 1424 |
+
"validation_metrics": "_valid_en_mlm_ppl,_valid_mlm_ppl,_valid_zh_mlm_ppl",
|
| 1425 |
+
"vocab_size": 200000,
|
| 1426 |
+
"word_blank": 0.0,
|
| 1427 |
+
"word_dropout": 0.0,
|
| 1428 |
+
"word_keep": 0.1,
|
| 1429 |
+
"word_mask": 0.8,
|
| 1430 |
+
"word_mask_keep_rand": "0.8,0.1,0.1",
|
| 1431 |
+
"word_pred": 0.15,
|
| 1432 |
+
"word_rand": 0.1,
|
| 1433 |
+
"word_shuffle": 0.0,
|
| 1434 |
+
"world_size": 32
|
| 1435 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"pad_token_id": 2,
|
| 5 |
+
"transformers_version": "4.34.0.dev0"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e9a140da6b8c24d5e43287bd6c17a174f7ea0b4ebf4f4158f698eb3235dfa22
|
| 3 |
+
size 297116
|
model.onnx_data
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42d2754b3ed45492dced18b2bd4dd00f3bccc9d59a12b3a943b9b2480ac9b172
|
| 3 |
+
size 2286787840
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<special0>",
|
| 4 |
+
"<special1>",
|
| 5 |
+
"<special2>",
|
| 6 |
+
"<special3>",
|
| 7 |
+
"<special4>",
|
| 8 |
+
"<special5>",
|
| 9 |
+
"<special6>",
|
| 10 |
+
"<special7>",
|
| 11 |
+
"<special8>",
|
| 12 |
+
"<special9>"
|
| 13 |
+
],
|
| 14 |
+
"bos_token": "<s>",
|
| 15 |
+
"cls_token": "</s>",
|
| 16 |
+
"mask_token": "<special1>",
|
| 17 |
+
"pad_token": "<pad>",
|
| 18 |
+
"sep_token": "</s>",
|
| 19 |
+
"unk_token": "<unk>"
|
| 20 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,228 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<special0>",
|
| 4 |
+
"<special1>",
|
| 5 |
+
"<special2>",
|
| 6 |
+
"<special3>",
|
| 7 |
+
"<special4>",
|
| 8 |
+
"<special5>",
|
| 9 |
+
"<special6>",
|
| 10 |
+
"<special7>",
|
| 11 |
+
"<special8>",
|
| 12 |
+
"<special9>"
|
| 13 |
+
],
|
| 14 |
+
"bos_token": "<s>",
|
| 15 |
+
"clean_up_tokenization_spaces": true,
|
| 16 |
+
"cls_token": "</s>",
|
| 17 |
+
"do_lowercase_and_remove_accent": false,
|
| 18 |
+
"id2lang": {
|
| 19 |
+
"0": "af",
|
| 20 |
+
"1": "als",
|
| 21 |
+
"10": "be",
|
| 22 |
+
"11": "bg",
|
| 23 |
+
"12": "bn",
|
| 24 |
+
"13": "br",
|
| 25 |
+
"14": "bs",
|
| 26 |
+
"15": "ca",
|
| 27 |
+
"16": "ceb",
|
| 28 |
+
"17": "ckb",
|
| 29 |
+
"18": "cs",
|
| 30 |
+
"19": "cy",
|
| 31 |
+
"2": "am",
|
| 32 |
+
"20": "da",
|
| 33 |
+
"21": "de",
|
| 34 |
+
"22": "el",
|
| 35 |
+
"23": "en",
|
| 36 |
+
"24": "eo",
|
| 37 |
+
"25": "es",
|
| 38 |
+
"26": "et",
|
| 39 |
+
"27": "eu",
|
| 40 |
+
"28": "fa",
|
| 41 |
+
"29": "fi",
|
| 42 |
+
"3": "an",
|
| 43 |
+
"30": "fr",
|
| 44 |
+
"31": "fy",
|
| 45 |
+
"32": "ga",
|
| 46 |
+
"33": "gan",
|
| 47 |
+
"34": "gl",
|
| 48 |
+
"35": "gu",
|
| 49 |
+
"36": "he",
|
| 50 |
+
"37": "hi",
|
| 51 |
+
"38": "hr",
|
| 52 |
+
"39": "hu",
|
| 53 |
+
"4": "ang",
|
| 54 |
+
"40": "hy",
|
| 55 |
+
"41": "ia",
|
| 56 |
+
"42": "id",
|
| 57 |
+
"43": "is",
|
| 58 |
+
"44": "it",
|
| 59 |
+
"45": "ja",
|
| 60 |
+
"46": "jv",
|
| 61 |
+
"47": "ka",
|
| 62 |
+
"48": "kk",
|
| 63 |
+
"49": "kn",
|
| 64 |
+
"5": "ar",
|
| 65 |
+
"50": "ko",
|
| 66 |
+
"51": "ku",
|
| 67 |
+
"52": "la",
|
| 68 |
+
"53": "lb",
|
| 69 |
+
"54": "lt",
|
| 70 |
+
"55": "lv",
|
| 71 |
+
"56": "mk",
|
| 72 |
+
"57": "ml",
|
| 73 |
+
"58": "mn",
|
| 74 |
+
"59": "mr",
|
| 75 |
+
"6": "arz",
|
| 76 |
+
"60": "ms",
|
| 77 |
+
"61": "my",
|
| 78 |
+
"62": "nds",
|
| 79 |
+
"63": "ne",
|
| 80 |
+
"64": "nl",
|
| 81 |
+
"65": "nn",
|
| 82 |
+
"66": "no",
|
| 83 |
+
"67": "oc",
|
| 84 |
+
"68": "pl",
|
| 85 |
+
"69": "pt",
|
| 86 |
+
"7": "ast",
|
| 87 |
+
"70": "ro",
|
| 88 |
+
"71": "ru",
|
| 89 |
+
"72": "scn",
|
| 90 |
+
"73": "sco",
|
| 91 |
+
"74": "sh",
|
| 92 |
+
"75": "si",
|
| 93 |
+
"76": "simple",
|
| 94 |
+
"77": "sk",
|
| 95 |
+
"78": "sl",
|
| 96 |
+
"79": "sq",
|
| 97 |
+
"8": "az",
|
| 98 |
+
"80": "sr",
|
| 99 |
+
"81": "sv",
|
| 100 |
+
"82": "sw",
|
| 101 |
+
"83": "ta",
|
| 102 |
+
"84": "te",
|
| 103 |
+
"85": "th",
|
| 104 |
+
"86": "tl",
|
| 105 |
+
"87": "tr",
|
| 106 |
+
"88": "tt",
|
| 107 |
+
"89": "uk",
|
| 108 |
+
"9": "bar",
|
| 109 |
+
"90": "ur",
|
| 110 |
+
"91": "uz",
|
| 111 |
+
"92": "vi",
|
| 112 |
+
"93": "war",
|
| 113 |
+
"94": "wuu",
|
| 114 |
+
"95": "yi",
|
| 115 |
+
"96": "zh",
|
| 116 |
+
"97": "zh_classical",
|
| 117 |
+
"98": "zh_min_nan",
|
| 118 |
+
"99": "zh_yue"
|
| 119 |
+
},
|
| 120 |
+
"lang2id": {
|
| 121 |
+
"af": 0,
|
| 122 |
+
"als": 1,
|
| 123 |
+
"am": 2,
|
| 124 |
+
"an": 3,
|
| 125 |
+
"ang": 4,
|
| 126 |
+
"ar": 5,
|
| 127 |
+
"arz": 6,
|
| 128 |
+
"ast": 7,
|
| 129 |
+
"az": 8,
|
| 130 |
+
"bar": 9,
|
| 131 |
+
"be": 10,
|
| 132 |
+
"bg": 11,
|
| 133 |
+
"bn": 12,
|
| 134 |
+
"br": 13,
|
| 135 |
+
"bs": 14,
|
| 136 |
+
"ca": 15,
|
| 137 |
+
"ceb": 16,
|
| 138 |
+
"ckb": 17,
|
| 139 |
+
"cs": 18,
|
| 140 |
+
"cy": 19,
|
| 141 |
+
"da": 20,
|
| 142 |
+
"de": 21,
|
| 143 |
+
"el": 22,
|
| 144 |
+
"en": 23,
|
| 145 |
+
"eo": 24,
|
| 146 |
+
"es": 25,
|
| 147 |
+
"et": 26,
|
| 148 |
+
"eu": 27,
|
| 149 |
+
"fa": 28,
|
| 150 |
+
"fi": 29,
|
| 151 |
+
"fr": 30,
|
| 152 |
+
"fy": 31,
|
| 153 |
+
"ga": 32,
|
| 154 |
+
"gan": 33,
|
| 155 |
+
"gl": 34,
|
| 156 |
+
"gu": 35,
|
| 157 |
+
"he": 36,
|
| 158 |
+
"hi": 37,
|
| 159 |
+
"hr": 38,
|
| 160 |
+
"hu": 39,
|
| 161 |
+
"hy": 40,
|
| 162 |
+
"ia": 41,
|
| 163 |
+
"id": 42,
|
| 164 |
+
"is": 43,
|
| 165 |
+
"it": 44,
|
| 166 |
+
"ja": 45,
|
| 167 |
+
"jv": 46,
|
| 168 |
+
"ka": 47,
|
| 169 |
+
"kk": 48,
|
| 170 |
+
"kn": 49,
|
| 171 |
+
"ko": 50,
|
| 172 |
+
"ku": 51,
|
| 173 |
+
"la": 52,
|
| 174 |
+
"lb": 53,
|
| 175 |
+
"lt": 54,
|
| 176 |
+
"lv": 55,
|
| 177 |
+
"mk": 56,
|
| 178 |
+
"ml": 57,
|
| 179 |
+
"mn": 58,
|
| 180 |
+
"mr": 59,
|
| 181 |
+
"ms": 60,
|
| 182 |
+
"my": 61,
|
| 183 |
+
"nds": 62,
|
| 184 |
+
"ne": 63,
|
| 185 |
+
"nl": 64,
|
| 186 |
+
"nn": 65,
|
| 187 |
+
"no": 66,
|
| 188 |
+
"oc": 67,
|
| 189 |
+
"pl": 68,
|
| 190 |
+
"pt": 69,
|
| 191 |
+
"ro": 70,
|
| 192 |
+
"ru": 71,
|
| 193 |
+
"scn": 72,
|
| 194 |
+
"sco": 73,
|
| 195 |
+
"sh": 74,
|
| 196 |
+
"si": 75,
|
| 197 |
+
"simple": 76,
|
| 198 |
+
"sk": 77,
|
| 199 |
+
"sl": 78,
|
| 200 |
+
"sq": 79,
|
| 201 |
+
"sr": 80,
|
| 202 |
+
"sv": 81,
|
| 203 |
+
"sw": 82,
|
| 204 |
+
"ta": 83,
|
| 205 |
+
"te": 84,
|
| 206 |
+
"th": 85,
|
| 207 |
+
"tl": 86,
|
| 208 |
+
"tr": 87,
|
| 209 |
+
"tt": 88,
|
| 210 |
+
"uk": 89,
|
| 211 |
+
"ur": 90,
|
| 212 |
+
"uz": 91,
|
| 213 |
+
"vi": 92,
|
| 214 |
+
"war": 93,
|
| 215 |
+
"wuu": 94,
|
| 216 |
+
"yi": 95,
|
| 217 |
+
"zh": 96,
|
| 218 |
+
"zh_classical": 97,
|
| 219 |
+
"zh_min_nan": 98,
|
| 220 |
+
"zh_yue": 99
|
| 221 |
+
},
|
| 222 |
+
"mask_token": "<special1>",
|
| 223 |
+
"model_max_length": 512,
|
| 224 |
+
"pad_token": "<pad>",
|
| 225 |
+
"sep_token": "</s>",
|
| 226 |
+
"tokenizer_class": "XLMTokenizer",
|
| 227 |
+
"unk_token": "<unk>"
|
| 228 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|