Wikidepia commited on
Commit
dc9995c
1 Parent(s): f1778bc

New model trained on 70M sentences

Browse files
config.json CHANGED
@@ -1,9 +1,6 @@
1
  {
2
- "_num_labels": 3,
3
  "activation_dropout": 0.0,
4
  "activation_function": "swish",
5
- "add_bias_logits": false,
6
- "add_final_layer_norm": false,
7
  "architectures": [
8
  "MarianMTModel"
9
  ],
@@ -14,7 +11,7 @@
14
  ]
15
  ],
16
  "bos_token_id": 0,
17
- "classif_dropout": 0.0,
18
  "d_model": 512,
19
  "decoder_attention_heads": 8,
20
  "decoder_ffn_dim": 2048,
@@ -27,27 +24,20 @@
27
  "encoder_layerdrop": 0.0,
28
  "encoder_layers": 6,
29
  "eos_token_id": 0,
30
- "id2label": {
31
- "0": "LABEL_0",
32
- "1": "LABEL_1",
33
- "2": "LABEL_2"
34
- },
35
  "init_std": 0.02,
36
  "is_encoder_decoder": true,
37
- "label2id": {
38
- "LABEL_0": 0,
39
- "LABEL_1": 1,
40
- "LABEL_2": 2
41
- },
42
  "max_length": 512,
43
  "max_position_embeddings": 512,
44
  "model_type": "marian",
45
- "normalize_before": false,
46
  "normalize_embedding": false,
47
- "num_beams": 4,
48
  "num_hidden_layers": 6,
49
  "pad_token_id": 32000,
50
  "scale_embedding": true,
51
- "static_position_embeddings": true,
 
 
52
  "vocab_size": 32001
53
  }
1
  {
 
2
  "activation_dropout": 0.0,
3
  "activation_function": "swish",
 
 
4
  "architectures": [
5
  "MarianMTModel"
6
  ],
11
  ]
12
  ],
13
  "bos_token_id": 0,
14
+ "classifier_dropout": 0.0,
15
  "d_model": 512,
16
  "decoder_attention_heads": 8,
17
  "decoder_ffn_dim": 2048,
24
  "encoder_layerdrop": 0.0,
25
  "encoder_layers": 6,
26
  "eos_token_id": 0,
27
+ "forced_eos_token_id": 0,
28
+ "gradient_checkpointing": false,
 
 
 
29
  "init_std": 0.02,
30
  "is_encoder_decoder": true,
 
 
 
 
 
31
  "max_length": 512,
32
  "max_position_embeddings": 512,
33
  "model_type": "marian",
 
34
  "normalize_embedding": false,
35
+ "num_beams": 6,
36
  "num_hidden_layers": 6,
37
  "pad_token_id": 32000,
38
  "scale_embedding": true,
39
+ "static_position_embeddings": false,
40
+ "transformers_version": "4.6.0",
41
+ "use_cache": true,
42
  "vocab_size": 32001
43
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9b95a1d5404bb1bebdafba2a2bb3380e76ebf79a76b75b763fe0ffd46827e946
3
- size 244368193
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:818d530007d20919ea20d2fd46bfcde6a8705da894fc405a071a4e91f4c79663
3
+ size 153963946
source.spm CHANGED
Binary files a/source.spm and b/source.spm differ
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
target.spm CHANGED
Binary files a/target.spm and b/target.spm differ
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"target_lang": "opusmt", "source_lang": ""}
1
+ {"source_lang": "", "target_lang": "npz", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "npz"}
vocab.json CHANGED
The diff for this file is too large to render. See raw diff