patrickvonplaten commited on
Commit
c48ad1f
1 Parent(s): 129775b

add tf model file

Browse files
Files changed (2) hide show
  1. config.json +9 -2
  2. tf_model.h5 +3 -0
config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "_num_labels": 3,
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
@@ -7,18 +8,23 @@
7
  "attention_dropout": 0.1,
8
  "bos_token_id": 0,
9
  "classif_dropout": 0.1,
 
10
  "d_model": 1024,
11
  "decoder_attention_heads": 16,
12
  "decoder_ffn_dim": 4096,
13
  "decoder_layerdrop": 0.0,
14
  "decoder_layers": 12,
15
- "dropout": 0.1,
 
 
16
  "encoder_attention_heads": 16,
17
  "encoder_ffn_dim": 4096,
18
  "encoder_layerdrop": 0.0,
19
  "encoder_layers": 12,
20
  "eos_token_id": 2,
21
  "extra_pos_embeddings": 2,
 
 
22
  "id2label": {
23
  "0": "LABEL_0",
24
  "1": "LABEL_1",
@@ -32,7 +38,6 @@
32
  "LABEL_2": 2
33
  },
34
  "max_length": 1024,
35
- "decoder_start_token_id": 250020,
36
  "max_position_embeddings": 1024,
37
  "model_type": "mbart",
38
  "normalize_before": true,
@@ -43,5 +48,7 @@
43
  "pad_token_id": 1,
44
  "scale_embedding": true,
45
  "static_position_embeddings": false,
 
 
46
  "vocab_size": 250027
47
  }
1
  {
2
+ "_name_or_path": "facebook/mbart-large-en-ro",
3
  "_num_labels": 3,
4
  "activation_dropout": 0.0,
5
  "activation_function": "gelu",
8
  "attention_dropout": 0.1,
9
  "bos_token_id": 0,
10
  "classif_dropout": 0.1,
11
+ "classifier_dropout": 0.0,
12
  "d_model": 1024,
13
  "decoder_attention_heads": 16,
14
  "decoder_ffn_dim": 4096,
15
  "decoder_layerdrop": 0.0,
16
  "decoder_layers": 12,
17
+ "decoder_start_token_id": 250020,
18
+ "do_blenderbot_90_layernorm": false,
19
+ "dropout": 0.1,
20
  "encoder_attention_heads": 16,
21
  "encoder_ffn_dim": 4096,
22
  "encoder_layerdrop": 0.0,
23
  "encoder_layers": 12,
24
  "eos_token_id": 2,
25
  "extra_pos_embeddings": 2,
26
+ "force_bos_token_to_be_generated": false,
27
+ "gradient_checkpointing": false,
28
  "id2label": {
29
  "0": "LABEL_0",
30
  "1": "LABEL_1",
38
  "LABEL_2": 2
39
  },
40
  "max_length": 1024,
 
41
  "max_position_embeddings": 1024,
42
  "model_type": "mbart",
43
  "normalize_before": true,
48
  "pad_token_id": 1,
49
  "scale_embedding": true,
50
  "static_position_embeddings": false,
51
+ "transformers_version": "4.2.0dev0",
52
+ "use_cache": true,
53
  "vocab_size": 250027
54
  }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51449b7a9d58d33245d5f69a87f7fd6b7c6becb5151dfff6bbd2b7d72d4b07eb
3
+ size 2443942072