Commit
•
9284f2b
1
Parent(s):
2194eea
update to pretraining
Browse files- config.json +10 -2
- pytorch_model.bin +2 -2
config.json
CHANGED
@@ -2,10 +2,12 @@
|
|
2 |
"activation_dropout": 0.1,
|
3 |
"apply_spec_augment": true,
|
4 |
"architectures": [
|
5 |
-
"
|
6 |
],
|
7 |
"attention_dropout": 0.1,
|
8 |
"bos_token_id": 1,
|
|
|
|
|
9 |
"conv_bias": false,
|
10 |
"conv_dim": [
|
11 |
512,
|
@@ -36,12 +38,14 @@
|
|
36 |
],
|
37 |
"ctc_loss_reduction": "sum",
|
38 |
"ctc_zero_infinity": false,
|
|
|
39 |
"do_stable_layer_norm": false,
|
40 |
"eos_token_id": 2,
|
41 |
"feat_extract_activation": "gelu",
|
42 |
"feat_extract_dropout": 0.0,
|
43 |
"feat_extract_norm": "group",
|
44 |
"feat_proj_dropout": 0.1,
|
|
|
45 |
"final_dropout": 0.1,
|
46 |
"gradient_checkpointing": false,
|
47 |
"hidden_act": "gelu",
|
@@ -58,11 +62,15 @@
|
|
58 |
"mask_time_prob": 0.05,
|
59 |
"model_type": "wav2vec2",
|
60 |
"num_attention_heads": 12,
|
|
|
|
|
61 |
"num_conv_pos_embedding_groups": 16,
|
62 |
"num_conv_pos_embeddings": 128,
|
63 |
"num_feat_extract_layers": 7,
|
64 |
"num_hidden_layers": 12,
|
|
|
65 |
"pad_token_id": 0,
|
66 |
-
"
|
|
|
67 |
"vocab_size": 32
|
68 |
}
|
|
|
2 |
"activation_dropout": 0.1,
|
3 |
"apply_spec_augment": true,
|
4 |
"architectures": [
|
5 |
+
"Wav2Vec2ForPreTraining"
|
6 |
],
|
7 |
"attention_dropout": 0.1,
|
8 |
"bos_token_id": 1,
|
9 |
+
"codevector_dim": 256,
|
10 |
+
"contrastive_logits_temperature": 0.1,
|
11 |
"conv_bias": false,
|
12 |
"conv_dim": [
|
13 |
512,
|
|
|
38 |
],
|
39 |
"ctc_loss_reduction": "sum",
|
40 |
"ctc_zero_infinity": false,
|
41 |
+
"diversity_loss_weight": 0.1,
|
42 |
"do_stable_layer_norm": false,
|
43 |
"eos_token_id": 2,
|
44 |
"feat_extract_activation": "gelu",
|
45 |
"feat_extract_dropout": 0.0,
|
46 |
"feat_extract_norm": "group",
|
47 |
"feat_proj_dropout": 0.1,
|
48 |
+
"feat_quantizer_dropout": 0.0,
|
49 |
"final_dropout": 0.1,
|
50 |
"gradient_checkpointing": false,
|
51 |
"hidden_act": "gelu",
|
|
|
62 |
"mask_time_prob": 0.05,
|
63 |
"model_type": "wav2vec2",
|
64 |
"num_attention_heads": 12,
|
65 |
+
"num_codevector_groups": 2,
|
66 |
+
"num_codevectors_per_group": 320,
|
67 |
"num_conv_pos_embedding_groups": 16,
|
68 |
"num_conv_pos_embeddings": 128,
|
69 |
"num_feat_extract_layers": 7,
|
70 |
"num_hidden_layers": 12,
|
71 |
+
"num_negatives": 100,
|
72 |
"pad_token_id": 0,
|
73 |
+
"proj_codevector_dim": 256,
|
74 |
+
"transformers_version": "4.7.0.dev0",
|
75 |
"vocab_size": 32
|
76 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0e0d07e954c8aad135e8ec3782ac26e2dd3e5fd97d136848f4791169e3023bf0
|
3 |
+
size 380267417
|