Hugging Face's logo

openpecha
/
wav2vec2-large-xlsr-tibetan

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Inference Endpoints
Model card Files Community
1
TenzinGayche commited on
Commit
45e517b
•
1 Parent(s): f4ca15a

Training in progress, step 300

Browse files
Files changed (19) hide show
  1. checkpoint-200/config.json +115 -0
  2. checkpoint-200/optimizer.pt +3 -0
  3. checkpoint-200/preprocessor_config.json +10 -0
  4. checkpoint-200/pytorch_model.bin +3 -0
  5. checkpoint-200/rng_state.pth +3 -0
  6. checkpoint-200/scaler.pt +3 -0
  7. checkpoint-200/scheduler.pt +3 -0
  8. checkpoint-200/trainer_state.json +58 -0
  9. checkpoint-200/training_args.bin +3 -0
  10. checkpoint-300/config.json +115 -0
  11. checkpoint-300/optimizer.pt +3 -0
  12. checkpoint-300/preprocessor_config.json +10 -0
  13. checkpoint-300/pytorch_model.bin +3 -0
  14. checkpoint-300/rng_state.pth +3 -0
  15. checkpoint-300/scaler.pt +3 -0
  16. checkpoint-300/scheduler.pt +3 -0
  17. checkpoint-300/trainer_state.json +79 -0
  18. checkpoint-300/training_args.bin +3 -0
  19. pytorch_model.bin +1 -1
checkpoint-200/config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ForCTC"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": true,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.0,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 1024,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 4096,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.1,
62
+ "mask_channel_length": 10,
63
+ "mask_channel_min_space": 1,
64
+ "mask_channel_other": 0.0,
65
+ "mask_channel_prob": 0.0,
66
+ "mask_channel_selection": "static",
67
+ "mask_feature_length": 10,
68
+ "mask_feature_min_masks": 0,
69
+ "mask_feature_prob": 0.0,
70
+ "mask_time_length": 10,
71
+ "mask_time_min_masks": 2,
72
+ "mask_time_min_space": 1,
73
+ "mask_time_other": 0.0,
74
+ "mask_time_prob": 0.05,
75
+ "mask_time_selection": "static",
76
+ "model_type": "wav2vec2",
77
+ "num_adapter_layers": 3,
78
+ "num_attention_heads": 16,
79
+ "num_codevector_groups": 2,
80
+ "num_codevectors_per_group": 320,
81
+ "num_conv_pos_embedding_groups": 16,
82
+ "num_conv_pos_embeddings": 128,
83
+ "num_feat_extract_layers": 7,
84
+ "num_hidden_layers": 24,
85
+ "num_negatives": 100,
86
+ "output_hidden_size": 1024,
87
+ "pad_token_id": 107,
88
+ "proj_codevector_dim": 768,
89
+ "tdnn_dilation": [
90
+ 1,
91
+ 2,
92
+ 3,
93
+ 1,
94
+ 1
95
+ ],
96
+ "tdnn_dim": [
97
+ 512,
98
+ 512,
99
+ 512,
100
+ 512,
101
+ 1500
102
+ ],
103
+ "tdnn_kernel": [
104
+ 5,
105
+ 3,
106
+ 3,
107
+ 1,
108
+ 1
109
+ ],
110
+ "torch_dtype": "float32",
111
+ "transformers_version": "4.29.2",
112
+ "use_weighted_layer_sum": false,
113
+ "vocab_size": 108,
114
+ "xvector_output_dim": 512
115
+ }
checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0653f90b453cbb96db095cf6a82917aa74b51748fa578f8c2668a6088aae19e6
3
+ size 2490946117
checkpoint-200/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
checkpoint-200/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54180f9199bd86a729cf967a8d46808d813c784c5f5d3301f00332fec949e8ce
3
+ size 1262344621
checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2a7e269384172f02323ddba2370ec68fb6f24fb5ab8dff488ba9b6e499abc70
3
+ size 14639
checkpoint-200/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:877f20fed985e54bb1a47fe2179a5dc18fe6bef951e08303cb939f5a948b87be
3
+ size 557
checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2d5d809d7f02c3427a0504e8958eeac273168e73eb818e70283f61f776acdd3
3
+ size 627
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.3605442176870748,
5
+ "global_step": 200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.34,
12
+ "learning_rate": 2.8199999999999998e-05,
13
+ "loss": 14.6495,
14
+ "step": 50
15
+ },
16
+ {
17
+ "epoch": 0.68,
18
+ "learning_rate": 5.82e-05,
19
+ "loss": 8.9314,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.68,
24
+ "eval_cer": 0.9940322843632806,
25
+ "eval_loss": 4.78579044342041,
26
+ "eval_runtime": 24.1442,
27
+ "eval_samples_per_second": 24.354,
28
+ "eval_steps_per_second": 3.065,
29
+ "step": 100
30
+ },
31
+ {
32
+ "epoch": 1.02,
33
+ "learning_rate": 8.819999999999999e-05,
34
+ "loss": 4.2225,
35
+ "step": 150
36
+ },
37
+ {
38
+ "epoch": 1.36,
39
+ "learning_rate": 0.0001182,
40
+ "loss": 3.2937,
41
+ "step": 200
42
+ },
43
+ {
44
+ "epoch": 1.36,
45
+ "eval_cer": 0.9940322843632806,
46
+ "eval_loss": 3.2489850521087646,
47
+ "eval_runtime": 24.3318,
48
+ "eval_samples_per_second": 24.166,
49
+ "eval_steps_per_second": 3.041,
50
+ "step": 200
51
+ }
52
+ ],
53
+ "max_steps": 3675,
54
+ "num_train_epochs": 25,
55
+ "total_flos": 8.003443015400256e+17,
56
+ "trial_name": null,
57
+ "trial_params": null
58
+ }
checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6818a2e9b105d4d459b9ef327417424b7dc79e7c93157b85684e997060ce05b5
3
+ size 3899
checkpoint-300/config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ForCTC"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": true,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "mean",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": true,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_dropout": 0.0,
51
+ "feat_extract_norm": "layer",
52
+ "feat_proj_dropout": 0.0,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 1024,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 4096,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.1,
62
+ "mask_channel_length": 10,
63
+ "mask_channel_min_space": 1,
64
+ "mask_channel_other": 0.0,
65
+ "mask_channel_prob": 0.0,
66
+ "mask_channel_selection": "static",
67
+ "mask_feature_length": 10,
68
+ "mask_feature_min_masks": 0,
69
+ "mask_feature_prob": 0.0,
70
+ "mask_time_length": 10,
71
+ "mask_time_min_masks": 2,
72
+ "mask_time_min_space": 1,
73
+ "mask_time_other": 0.0,
74
+ "mask_time_prob": 0.05,
75
+ "mask_time_selection": "static",
76
+ "model_type": "wav2vec2",
77
+ "num_adapter_layers": 3,
78
+ "num_attention_heads": 16,
79
+ "num_codevector_groups": 2,
80
+ "num_codevectors_per_group": 320,
81
+ "num_conv_pos_embedding_groups": 16,
82
+ "num_conv_pos_embeddings": 128,
83
+ "num_feat_extract_layers": 7,
84
+ "num_hidden_layers": 24,
85
+ "num_negatives": 100,
86
+ "output_hidden_size": 1024,
87
+ "pad_token_id": 107,
88
+ "proj_codevector_dim": 768,
89
+ "tdnn_dilation": [
90
+ 1,
91
+ 2,
92
+ 3,
93
+ 1,
94
+ 1
95
+ ],
96
+ "tdnn_dim": [
97
+ 512,
98
+ 512,
99
+ 512,
100
+ 512,
101
+ 1500
102
+ ],
103
+ "tdnn_kernel": [
104
+ 5,
105
+ 3,
106
+ 3,
107
+ 1,
108
+ 1
109
+ ],
110
+ "torch_dtype": "float32",
111
+ "transformers_version": "4.29.2",
112
+ "use_weighted_layer_sum": false,
113
+ "vocab_size": 108,
114
+ "xvector_output_dim": 512
115
+ }
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24c6185292f6deccbd0f53f7d29f70b969775bcd45bab7f1bd68f92a58caa41c
3
+ size 2490946501
checkpoint-300/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
checkpoint-300/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:824b2e6010cca74450a56443d3c7d7bef2cffb1fba986581c458d3f71cf165c6
3
+ size 1262344621
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee6ea403da48fb17774ad02b6b8ac9d9fbe59fd9baa346ca66a7aa460b490c8b
3
+ size 14639
checkpoint-300/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b4a0fdc001d19ad30572053968b78c78dfa2738dfe7692e61c0ec5df4f49902
3
+ size 557
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23b261d75cef5703f8bbd243e8e8b0e2ac902586109285db80db5c16affdbece
3
+ size 627
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0408163265306123,
5
+ "global_step": 300,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.34,
12
+ "learning_rate": 2.8199999999999998e-05,
13
+ "loss": 14.6495,
14
+ "step": 50
15
+ },
16
+ {
17
+ "epoch": 0.68,
18
+ "learning_rate": 5.82e-05,
19
+ "loss": 8.9314,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.68,
24
+ "eval_cer": 0.9940322843632806,
25
+ "eval_loss": 4.78579044342041,
26
+ "eval_runtime": 24.1442,
27
+ "eval_samples_per_second": 24.354,
28
+ "eval_steps_per_second": 3.065,
29
+ "step": 100
30
+ },
31
+ {
32
+ "epoch": 1.02,
33
+ "learning_rate": 8.819999999999999e-05,
34
+ "loss": 4.2225,
35
+ "step": 150
36
+ },
37
+ {
38
+ "epoch": 1.36,
39
+ "learning_rate": 0.0001182,
40
+ "loss": 3.2937,
41
+ "step": 200
42
+ },
43
+ {
44
+ "epoch": 1.36,
45
+ "eval_cer": 0.9940322843632806,
46
+ "eval_loss": 3.2489850521087646,
47
+ "eval_runtime": 24.3318,
48
+ "eval_samples_per_second": 24.166,
49
+ "eval_steps_per_second": 3.041,
50
+ "step": 200
51
+ },
52
+ {
53
+ "epoch": 1.7,
54
+ "learning_rate": 0.0001482,
55
+ "loss": 3.2508,
56
+ "step": 250
57
+ },
58
+ {
59
+ "epoch": 2.04,
60
+ "learning_rate": 0.00017819999999999997,
61
+ "loss": 3.2412,
62
+ "step": 300
63
+ },
64
+ {
65
+ "epoch": 2.04,
66
+ "eval_cer": 0.9940322843632806,
67
+ "eval_loss": 3.222844123840332,
68
+ "eval_runtime": 24.3579,
69
+ "eval_samples_per_second": 24.14,
70
+ "eval_steps_per_second": 3.038,
71
+ "step": 300
72
+ }
73
+ ],
74
+ "max_steps": 3675,
75
+ "num_train_epochs": 25,
76
+ "total_flos": 1.1789672542468416e+18,
77
+ "trial_name": null,
78
+ "trial_params": null
79
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6818a2e9b105d4d459b9ef327417424b7dc79e7c93157b85684e997060ce05b5
3
+ size 3899
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:48d0e78d44dcb13959b6477b9dbef229431b5ab93be05782631cd39e19428348
3
  size 1262344621
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:824b2e6010cca74450a56443d3c7d7bef2cffb1fba986581c458d3f71cf165c6
3
  size 1262344621