Hugging Face's logo

jssaluja
/
fb-mms-1b-cleaned-jssaluja_rajinder_singh-epochs-12-test-datasets-10-20250812_232950

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Panjabi
wav2vec2
hf-asr-leaderboard
Generated from Trainer
Model card Files
xet
Metrics Community
jssaluja commited on
Commit
d245eeb
·
verified ·
1 Parent(s): 47739cc

Training in progress, epoch 10, checkpoint

Browse files
Files changed (9) hide show
  1. checkpoint-75480/config.json +107 -0
  2. checkpoint-75480/model.safetensors +3 -0
  3. checkpoint-75480/optimizer.pt +3 -0
  4. checkpoint-75480/preprocessor_config.json +9 -0
  5. checkpoint-75480/rng_state.pth +3 -0
  6. checkpoint-75480/scaler.pt +3 -0
  7. checkpoint-75480/scheduler.pt +3 -0
  8. checkpoint-75480/trainer_state.json +224 -0
  9. checkpoint-75480/training_args.bin +3 -0
checkpoint-75480/config.json ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.05,
3
+ "adapter_attn_dim": 16,
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": 0.0,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 1024,
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": 0.05,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.0,
57
+ "hidden_size": 1280,
58
+ "initializer_range": 0.02,
59
+ "intermediate_size": 5120,
60
+ "layer_norm_eps": 1e-05,
61
+ "layerdrop": 0.0,
62
+ "mask_feature_length": 10,
63
+ "mask_feature_min_masks": 0,
64
+ "mask_feature_prob": 0.0,
65
+ "mask_time_length": 10,
66
+ "mask_time_min_masks": 2,
67
+ "mask_time_prob": 0.05,
68
+ "model_type": "wav2vec2",
69
+ "num_adapter_layers": 3,
70
+ "num_attention_heads": 16,
71
+ "num_codevector_groups": 2,
72
+ "num_codevectors_per_group": 320,
73
+ "num_conv_pos_embedding_groups": 16,
74
+ "num_conv_pos_embeddings": 128,
75
+ "num_feat_extract_layers": 7,
76
+ "num_hidden_layers": 48,
77
+ "num_negatives": 100,
78
+ "output_hidden_size": 1280,
79
+ "pad_token_id": 62,
80
+ "proj_codevector_dim": 1024,
81
+ "tdnn_dilation": [
82
+ 1,
83
+ 2,
84
+ 3,
85
+ 1,
86
+ 1
87
+ ],
88
+ "tdnn_dim": [
89
+ 512,
90
+ 512,
91
+ 512,
92
+ 512,
93
+ 1500
94
+ ],
95
+ "tdnn_kernel": [
96
+ 5,
97
+ 3,
98
+ 3,
99
+ 1,
100
+ 1
101
+ ],
102
+ "torch_dtype": "float32",
103
+ "transformers_version": "4.55.0",
104
+ "use_weighted_layer_sum": false,
105
+ "vocab_size": 65,
106
+ "xvector_output_dim": 512
107
+ }
checkpoint-75480/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c6304c45078627a002ac1786cacc7a8116fcab7e5127f2b5067efc034b4bf62
3
+ size 3859065140
checkpoint-75480/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9625801f5404d6e7a3b243db394bad34a58e40265809cf2e9f0dab6087aa04f3
3
+ size 18116858
checkpoint-75480/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-75480/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33b734299378c46e86b55be37330d55eeaa958b9002a7a582433d638974d4570
3
+ size 14308
checkpoint-75480/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fce8c8b59bbb81f69ef9cf6b35928cd2adf43aeb53383b423a6682d70c04f0e1
3
+ size 988
checkpoint-75480/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0d980ecbadd5666f7a0ac9d2193ce6b35c0c30966638b6470129c6f625e9305
3
+ size 1064
checkpoint-75480/trainer_state.json ADDED
@@ -0,0 +1,224 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 75480,
3
+ "best_metric": 0.35844873332606403,
4
+ "best_model_checkpoint": "fb-mms-1b-cleaned-jssaluja_rajinder_singh-epochs-12-test-datasets-10-20250812_232950/checkpoint-75480",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 75480,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 0.42587602138519287,
15
+ "learning_rate": 0.0009188058776666371,
16
+ "loss": 0.387,
17
+ "step": 7548
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_cer": 0.12580931434965467,
22
+ "eval_loss": 0.4400504529476166,
23
+ "eval_mer": 0.4452148370790355,
24
+ "eval_runtime": 175.5013,
25
+ "eval_samples_per_second": 8.94,
26
+ "eval_steps_per_second": 1.122,
27
+ "eval_wer": 0.4737760331479664,
28
+ "eval_wil": 0.6682418758522689,
29
+ "step": 7548
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "grad_norm": 2.229217290878296,
34
+ "learning_rate": 0.0008353213242453749,
35
+ "loss": 0.2054,
36
+ "step": 15096
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_cer": 0.10598448894547981,
41
+ "eval_loss": 0.3444209694862366,
42
+ "eval_mer": 0.3981059527200228,
43
+ "eval_runtime": 176.4752,
44
+ "eval_samples_per_second": 8.891,
45
+ "eval_steps_per_second": 1.116,
46
+ "eval_wer": 0.4064260531385163,
47
+ "eval_wil": 0.6145206422660318,
48
+ "step": 15096
49
+ },
50
+ {
51
+ "epoch": 3.0,
52
+ "grad_norm": 1.2057112455368042,
53
+ "learning_rate": 0.0007501545663310369,
54
+ "loss": 0.1811,
55
+ "step": 22644
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_cer": 0.11019793957633986,
60
+ "eval_loss": 0.3718140423297882,
61
+ "eval_mer": 0.4041122692184554,
62
+ "eval_runtime": 117.258,
63
+ "eval_samples_per_second": 13.381,
64
+ "eval_steps_per_second": 1.68,
65
+ "eval_wer": 0.4128957220223167,
66
+ "eval_wil": 0.6234012866542855,
67
+ "step": 22644
68
+ },
69
+ {
70
+ "epoch": 4.0,
71
+ "grad_norm": 0.06720034778118134,
72
+ "learning_rate": 0.0006668433139021374,
73
+ "loss": 0.1666,
74
+ "step": 30192
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_cer": 0.10207971601651426,
79
+ "eval_loss": 0.33335745334625244,
80
+ "eval_mer": 0.39927228186468766,
81
+ "eval_runtime": 116.5053,
82
+ "eval_samples_per_second": 13.467,
83
+ "eval_steps_per_second": 1.691,
84
+ "eval_wer": 0.40282775415258243,
85
+ "eval_wil": 0.6107857931270173,
86
+ "step": 30192
87
+ },
88
+ {
89
+ "epoch": 5.0,
90
+ "grad_norm": 1.2346702814102173,
91
+ "learning_rate": 0.0005835431019254549,
92
+ "loss": 0.1548,
93
+ "step": 37740
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_cer": 0.09676274260138133,
98
+ "eval_loss": 0.32630863785743713,
99
+ "eval_mer": 0.36752806986276954,
100
+ "eval_runtime": 118.542,
101
+ "eval_samples_per_second": 13.236,
102
+ "eval_steps_per_second": 1.662,
103
+ "eval_wer": 0.3747682913531785,
104
+ "eval_wil": 0.5768814499686782,
105
+ "step": 37740
106
+ },
107
+ {
108
+ "epoch": 6.0,
109
+ "grad_norm": 0.6912559270858765,
110
+ "learning_rate": 0.0005002428899487724,
111
+ "loss": 0.1452,
112
+ "step": 45288
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_cer": 0.09863024269784311,
117
+ "eval_loss": 0.3266763687133789,
118
+ "eval_mer": 0.3836833021717244,
119
+ "eval_runtime": 116.6891,
120
+ "eval_samples_per_second": 13.446,
121
+ "eval_steps_per_second": 1.688,
122
+ "eval_wer": 0.38785301493839275,
123
+ "eval_wil": 0.5942902200119844,
124
+ "step": 45288
125
+ },
126
+ {
127
+ "epoch": 7.0,
128
+ "grad_norm": 0.30876782536506653,
129
+ "learning_rate": 0.0004169316375198728,
130
+ "loss": 0.1382,
131
+ "step": 52836
132
+ },
133
+ {
134
+ "epoch": 7.0,
135
+ "eval_cer": 0.09598333140409769,
136
+ "eval_loss": 0.3241690993309021,
137
+ "eval_mer": 0.37248430493273543,
138
+ "eval_runtime": 118.5252,
139
+ "eval_samples_per_second": 13.238,
140
+ "eval_steps_per_second": 1.662,
141
+ "eval_wer": 0.37738523607022134,
142
+ "eval_wil": 0.5809274982110599,
143
+ "step": 52836
144
+ },
145
+ {
146
+ "epoch": 8.0,
147
+ "grad_norm": 0.05797579139471054,
148
+ "learning_rate": 0.0003336203850909733,
149
+ "loss": 0.1314,
150
+ "step": 60384
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_cer": 0.094926110275109,
155
+ "eval_loss": 0.3272537887096405,
156
+ "eval_mer": 0.3580110889963037,
157
+ "eval_runtime": 118.3389,
158
+ "eval_samples_per_second": 13.259,
159
+ "eval_steps_per_second": 1.665,
160
+ "eval_wer": 0.3661178352051757,
161
+ "eval_wil": 0.5655788246627147,
162
+ "step": 60384
163
+ },
164
+ {
165
+ "epoch": 9.0,
166
+ "grad_norm": 0.2719763517379761,
167
+ "learning_rate": 0.0002503201731142908,
168
+ "loss": 0.1257,
169
+ "step": 67932
170
+ },
171
+ {
172
+ "epoch": 9.0,
173
+ "eval_cer": 0.0944708106648146,
174
+ "eval_loss": 0.32030296325683594,
175
+ "eval_mer": 0.36196690914277563,
176
+ "eval_runtime": 119.7609,
177
+ "eval_samples_per_second": 13.101,
178
+ "eval_steps_per_second": 1.645,
179
+ "eval_wer": 0.3689528586486388,
180
+ "eval_wil": 0.5697933078898477,
181
+ "step": 67932
182
+ },
183
+ {
184
+ "epoch": 10.0,
185
+ "grad_norm": 0.3732568919658661,
186
+ "learning_rate": 0.00016700892068539127,
187
+ "loss": 0.1195,
188
+ "step": 75480
189
+ },
190
+ {
191
+ "epoch": 10.0,
192
+ "eval_cer": 0.0921094262453216,
193
+ "eval_loss": 0.31533417105674744,
194
+ "eval_mer": 0.35279387565285825,
195
+ "eval_runtime": 119.2627,
196
+ "eval_samples_per_second": 13.156,
197
+ "eval_steps_per_second": 1.652,
198
+ "eval_wer": 0.35844873332606403,
199
+ "eval_wil": 0.5575537089491287,
200
+ "step": 75480
201
+ }
202
+ ],
203
+ "logging_steps": 500,
204
+ "max_steps": 90576,
205
+ "num_input_tokens_seen": 0,
206
+ "num_train_epochs": 12,
207
+ "save_steps": 500,
208
+ "stateful_callbacks": {
209
+ "TrainerControl": {
210
+ "args": {
211
+ "should_epoch_stop": false,
212
+ "should_evaluate": false,
213
+ "should_log": false,
214
+ "should_save": true,
215
+ "should_training_stop": false
216
+ },
217
+ "attributes": {}
218
+ }
219
+ },
220
+ "total_flos": 2.7656116671414718e+20,
221
+ "train_batch_size": 8,
222
+ "trial_name": null,
223
+ "trial_params": null
224
+ }
checkpoint-75480/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d69e65a25bdaa9fcab13cdcd94b922f6df8111f5368fadc209379e61dda3af48
3
+ size 5624