Commit
·
8ffcd27
1
Parent(s):
dabc149
v2 update
Browse files- config.json +2 -4
- merges.txt +0 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- vocab-src.json +0 -0
- vocab-tgt.json +0 -0
config.json
CHANGED
|
@@ -59,7 +59,7 @@
|
|
| 59 |
"top_p": 1.0,
|
| 60 |
"torch_dtype": null,
|
| 61 |
"torchscript": false,
|
| 62 |
-
"transformers_version": "4.
|
| 63 |
"use_bfloat16": false,
|
| 64 |
"vocab_size": 99
|
| 65 |
},
|
|
@@ -81,10 +81,8 @@
|
|
| 81 |
"ru",
|
| 82 |
"en"
|
| 83 |
],
|
| 84 |
-
"max_length": 200,
|
| 85 |
"max_position_embeddings": 20,
|
| 86 |
"model_type": "fsmt",
|
| 87 |
-
"num_beams": 5,
|
| 88 |
"num_hidden_layers": 2,
|
| 89 |
"pad_token_id": 1,
|
| 90 |
"scale_embedding": true,
|
|
@@ -93,5 +91,5 @@
|
|
| 93 |
"tie_word_embeddings": false,
|
| 94 |
"transformers_version": null,
|
| 95 |
"use_cache": true,
|
| 96 |
-
"vocab_size":
|
| 97 |
}
|
|
|
|
| 59 |
"top_p": 1.0,
|
| 60 |
"torch_dtype": null,
|
| 61 |
"torchscript": false,
|
| 62 |
+
"transformers_version": "4.11.0.dev0",
|
| 63 |
"use_bfloat16": false,
|
| 64 |
"vocab_size": 99
|
| 65 |
},
|
|
|
|
| 81 |
"ru",
|
| 82 |
"en"
|
| 83 |
],
|
|
|
|
| 84 |
"max_position_embeddings": 20,
|
| 85 |
"model_type": "fsmt",
|
|
|
|
| 86 |
"num_hidden_layers": 2,
|
| 87 |
"pad_token_id": 1,
|
| 88 |
"scale_embedding": true,
|
|
|
|
| 91 |
"tie_word_embeddings": false,
|
| 92 |
"transformers_version": null,
|
| 93 |
"use_cache": true,
|
| 94 |
+
"vocab_size": 31232
|
| 95 |
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"langs": ["ru", "en"], "do_lower_case": false, "unk_token": "<unk>", "bos_token": "<s>", "sep_token": "</s>", "pad_token": "<pad>", "model_max_length": 1024, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "facebook/wmt19-ru-en", "tokenizer_class": "FSMTTokenizer"}
|
vocab-src.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
vocab-tgt.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|