Hugging Face's logo

spktsagar
/
wav2vec2-large-xls-r-300m-nepali-openslr

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Nepali
wav2vec2
Generated from Trainer
speech
openslr
nepali
Eval Results
Inference Endpoints
Model card Files Metrics Community
1
spktsagar commited on
Commit
46ab135
•
1 Parent(s): af12739

add checkpoint

Browse files
Files changed (9) hide show
  1. checkpoint-17600/config.json +108 -0
  2. checkpoint-17600/optimizer.pt +3 -0
  3. checkpoint-17600/preprocessor_config.json +10 -0
  4. checkpoint-17600/pytorch_model.bin +3 -0
  5. checkpoint-17600/rng_state.pth +3 -0
  6. checkpoint-17600/scaler.pt +3 -0
  7. checkpoint-17600/scheduler.pt +3 -0
  8. checkpoint-17600/trainer_state.json +346 -0
  9. checkpoint-17600/training_args.bin +3 -0
checkpoint-17600/config.json ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-xls-r-300m",
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
+ "gradient_checkpointing": false,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_size": 1024,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 4096,
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.1,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.075,
69
+ "model_type": "wav2vec2",
70
+ "num_adapter_layers": 3,
71
+ "num_attention_heads": 16,
72
+ "num_codevector_groups": 2,
73
+ "num_codevectors_per_group": 320,
74
+ "num_conv_pos_embedding_groups": 16,
75
+ "num_conv_pos_embeddings": 128,
76
+ "num_feat_extract_layers": 7,
77
+ "num_hidden_layers": 24,
78
+ "num_negatives": 100,
79
+ "output_hidden_size": 1024,
80
+ "pad_token_id": 0,
81
+ "proj_codevector_dim": 768,
82
+ "tdnn_dilation": [
83
+ 1,
84
+ 2,
85
+ 3,
86
+ 1,
87
+ 1
88
+ ],
89
+ "tdnn_dim": [
90
+ 512,
91
+ 512,
92
+ 512,
93
+ 512,
94
+ 1500
95
+ ],
96
+ "tdnn_kernel": [
97
+ 5,
98
+ 3,
99
+ 3,
100
+ 1,
101
+ 1
102
+ ],
103
+ "torch_dtype": "float32",
104
+ "transformers_version": "4.23.1",
105
+ "use_weighted_layer_sum": false,
106
+ "vocab_size": 81,
107
+ "xvector_output_dim": 512
108
+ }
checkpoint-17600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33663b2e32bd2c7c9b33a624c0579968fc626124e80624ae28686b30470d1f58
3
+ size 2490723217
checkpoint-17600/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-17600/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8741d8eb5349e35ea5c546ba44ea87721ee8471eff13aa064ac1baa2951cbc0
3
+ size 1262230833
checkpoint-17600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f9f6c5878a6d847a85918d7cf7d6178f6d32da5153191da2074de5b45e26714
3
+ size 14567
checkpoint-17600/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196d38a71c67e36d6c64325fbdcb8a1048cbfb9ae04a6c7478207cc7b21c0fa3
3
+ size 559
checkpoint-17600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:834eed3762ff9d3b3cb635c2055528468cdd7241dd343a74e85bf9d54728f4ea
3
+ size 623
checkpoint-17600/trainer_state.json ADDED
@@ -0,0 +1,346 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.602431690417048,
5
+ "global_step": 17600,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.21,
12
+ "learning_rate": 0.0002976391096979332,
13
+ "loss": 6.4525,
14
+ "step": 800
15
+ },
16
+ {
17
+ "epoch": 0.21,
18
+ "eval_loss": 1.4264159202575684,
19
+ "eval_runtime": 579.6055,
20
+ "eval_samples_per_second": 37.262,
21
+ "eval_steps_per_second": 4.658,
22
+ "eval_wer": 0.9522086267891338,
23
+ "step": 800
24
+ },
25
+ {
26
+ "epoch": 0.42,
27
+ "learning_rate": 0.00029127980922098566,
28
+ "loss": 1.105,
29
+ "step": 1600
30
+ },
31
+ {
32
+ "epoch": 0.42,
33
+ "eval_loss": 0.5966897010803223,
34
+ "eval_runtime": 638.739,
35
+ "eval_samples_per_second": 33.812,
36
+ "eval_steps_per_second": 4.227,
37
+ "eval_wer": 0.6514394209860115,
38
+ "step": 1600
39
+ },
40
+ {
41
+ "epoch": 0.63,
42
+ "learning_rate": 0.00028492050874403814,
43
+ "loss": 0.7933,
44
+ "step": 2400
45
+ },
46
+ {
47
+ "epoch": 0.63,
48
+ "eval_loss": 0.44666677713394165,
49
+ "eval_runtime": 578.27,
50
+ "eval_samples_per_second": 37.348,
51
+ "eval_steps_per_second": 4.669,
52
+ "eval_wer": 0.5045276037778715,
53
+ "step": 2400
54
+ },
55
+ {
56
+ "epoch": 0.84,
57
+ "learning_rate": 0.00027856120826709056,
58
+ "loss": 0.6926,
59
+ "step": 3200
60
+ },
61
+ {
62
+ "epoch": 0.84,
63
+ "eval_loss": 0.3889370858669281,
64
+ "eval_runtime": 574.1787,
65
+ "eval_samples_per_second": 37.614,
66
+ "eval_steps_per_second": 4.702,
67
+ "eval_wer": 0.4508941611762033,
68
+ "step": 3200
69
+ },
70
+ {
71
+ "epoch": 1.05,
72
+ "learning_rate": 0.0002722019077901431,
73
+ "loss": 0.6258,
74
+ "step": 4000
75
+ },
76
+ {
77
+ "epoch": 1.05,
78
+ "eval_loss": 0.33456486463546753,
79
+ "eval_runtime": 575.0581,
80
+ "eval_samples_per_second": 37.556,
81
+ "eval_steps_per_second": 4.695,
82
+ "eval_wer": 0.39778001363149523,
83
+ "step": 4000
84
+ },
85
+ {
86
+ "epoch": 1.26,
87
+ "learning_rate": 0.0002658426073131955,
88
+ "loss": 0.5666,
89
+ "step": 4800
90
+ },
91
+ {
92
+ "epoch": 1.26,
93
+ "eval_loss": 0.31256240606307983,
94
+ "eval_runtime": 592.8115,
95
+ "eval_samples_per_second": 36.431,
96
+ "eval_steps_per_second": 4.555,
97
+ "eval_wer": 0.3761481289149979,
98
+ "step": 4800
99
+ },
100
+ {
101
+ "epoch": 1.46,
102
+ "learning_rate": 0.000259483306836248,
103
+ "loss": 0.5501,
104
+ "step": 5600
105
+ },
106
+ {
107
+ "epoch": 1.46,
108
+ "eval_loss": 0.2980588376522064,
109
+ "eval_runtime": 662.1285,
110
+ "eval_samples_per_second": 32.618,
111
+ "eval_steps_per_second": 4.078,
112
+ "eval_wer": 0.3593684073869722,
113
+ "step": 5600
114
+ },
115
+ {
116
+ "epoch": 1.67,
117
+ "learning_rate": 0.00025313195548489663,
118
+ "loss": 0.5275,
119
+ "step": 6400
120
+ },
121
+ {
122
+ "epoch": 1.67,
123
+ "eval_loss": 0.28137603402137756,
124
+ "eval_runtime": 588.9138,
125
+ "eval_samples_per_second": 36.673,
126
+ "eval_steps_per_second": 4.585,
127
+ "eval_wer": 0.3457206841712375,
128
+ "step": 6400
129
+ },
130
+ {
131
+ "epoch": 1.88,
132
+ "learning_rate": 0.0002467726550079491,
133
+ "loss": 0.5203,
134
+ "step": 7200
135
+ },
136
+ {
137
+ "epoch": 1.88,
138
+ "eval_loss": 0.2654956877231598,
139
+ "eval_runtime": 586.4688,
140
+ "eval_samples_per_second": 36.825,
141
+ "eval_steps_per_second": 4.604,
142
+ "eval_wer": 0.3303041121677323,
143
+ "step": 7200
144
+ },
145
+ {
146
+ "epoch": 2.09,
147
+ "learning_rate": 0.00024041335453100157,
148
+ "loss": 0.4842,
149
+ "step": 8000
150
+ },
151
+ {
152
+ "epoch": 2.09,
153
+ "eval_loss": 0.2574324607849121,
154
+ "eval_runtime": 641.5915,
155
+ "eval_samples_per_second": 33.662,
156
+ "eval_steps_per_second": 4.208,
157
+ "eval_wer": 0.31566648274966735,
158
+ "step": 8000
159
+ },
160
+ {
161
+ "epoch": 2.3,
162
+ "learning_rate": 0.00023405405405405404,
163
+ "loss": 0.4542,
164
+ "step": 8800
165
+ },
166
+ {
167
+ "epoch": 2.3,
168
+ "eval_loss": 0.2577317953109741,
169
+ "eval_runtime": 583.455,
170
+ "eval_samples_per_second": 37.016,
171
+ "eval_steps_per_second": 4.628,
172
+ "eval_wer": 0.31417350946090683,
173
+ "step": 8800
174
+ },
175
+ {
176
+ "epoch": 2.51,
177
+ "learning_rate": 0.0002276947535771065,
178
+ "loss": 0.4431,
179
+ "step": 9600
180
+ },
181
+ {
182
+ "epoch": 2.51,
183
+ "eval_loss": 0.23904646933078766,
184
+ "eval_runtime": 581.9022,
185
+ "eval_samples_per_second": 37.114,
186
+ "eval_steps_per_second": 4.64,
187
+ "eval_wer": 0.2938236344162799,
188
+ "step": 9600
189
+ },
190
+ {
191
+ "epoch": 2.72,
192
+ "learning_rate": 0.00022133545310015898,
193
+ "loss": 0.4235,
194
+ "step": 10400
195
+ },
196
+ {
197
+ "epoch": 2.72,
198
+ "eval_loss": 0.24016053974628448,
199
+ "eval_runtime": 582.7306,
200
+ "eval_samples_per_second": 37.062,
201
+ "eval_steps_per_second": 4.633,
202
+ "eval_wer": 0.2884684041413781,
203
+ "step": 10400
204
+ },
205
+ {
206
+ "epoch": 2.93,
207
+ "learning_rate": 0.0002149920508744038,
208
+ "loss": 0.4202,
209
+ "step": 11200
210
+ },
211
+ {
212
+ "epoch": 2.93,
213
+ "eval_loss": 0.22816315293312073,
214
+ "eval_runtime": 643.2534,
215
+ "eval_samples_per_second": 33.575,
216
+ "eval_steps_per_second": 4.197,
217
+ "eval_wer": 0.2846710590373568,
218
+ "step": 11200
219
+ },
220
+ {
221
+ "epoch": 3.14,
222
+ "learning_rate": 0.00020863275039745627,
223
+ "loss": 0.4018,
224
+ "step": 12000
225
+ },
226
+ {
227
+ "epoch": 3.14,
228
+ "eval_loss": 0.22759725153446198,
229
+ "eval_runtime": 584.1242,
230
+ "eval_samples_per_second": 36.973,
231
+ "eval_steps_per_second": 4.622,
232
+ "eval_wer": 0.27702768491772417,
233
+ "step": 12000
234
+ },
235
+ {
236
+ "epoch": 3.35,
237
+ "learning_rate": 0.00020227344992050872,
238
+ "loss": 0.3986,
239
+ "step": 12800
240
+ },
241
+ {
242
+ "epoch": 3.35,
243
+ "eval_loss": 0.224422425031662,
244
+ "eval_runtime": 582.249,
245
+ "eval_samples_per_second": 37.092,
246
+ "eval_steps_per_second": 4.637,
247
+ "eval_wer": 0.27022816526565185,
248
+ "step": 12800
249
+ },
250
+ {
251
+ "epoch": 3.56,
252
+ "learning_rate": 0.00019591414944356117,
253
+ "loss": 0.3792,
254
+ "step": 13600
255
+ },
256
+ {
257
+ "epoch": 3.56,
258
+ "eval_loss": 0.21611596643924713,
259
+ "eval_runtime": 589.0669,
260
+ "eval_samples_per_second": 36.663,
261
+ "eval_steps_per_second": 4.584,
262
+ "eval_wer": 0.2742527019570932,
263
+ "step": 13600
264
+ },
265
+ {
266
+ "epoch": 3.77,
267
+ "learning_rate": 0.00018955484896661365,
268
+ "loss": 0.3736,
269
+ "step": 14400
270
+ },
271
+ {
272
+ "epoch": 3.77,
273
+ "eval_loss": 0.20533400774002075,
274
+ "eval_runtime": 644.3789,
275
+ "eval_samples_per_second": 33.516,
276
+ "eval_steps_per_second": 4.19,
277
+ "eval_wer": 0.2546979974684366,
278
+ "step": 14400
279
+ },
280
+ {
281
+ "epoch": 3.97,
282
+ "learning_rate": 0.0001831955484896661,
283
+ "loss": 0.37,
284
+ "step": 15200
285
+ },
286
+ {
287
+ "epoch": 3.97,
288
+ "eval_loss": 0.19958257675170898,
289
+ "eval_runtime": 585.9573,
290
+ "eval_samples_per_second": 36.858,
291
+ "eval_steps_per_second": 4.608,
292
+ "eval_wer": 0.25521729252539677,
293
+ "step": 15200
294
+ },
295
+ {
296
+ "epoch": 4.18,
297
+ "learning_rate": 0.00017683624801271858,
298
+ "loss": 0.3499,
299
+ "step": 16000
300
+ },
301
+ {
302
+ "epoch": 4.18,
303
+ "eval_loss": 0.19865886867046356,
304
+ "eval_runtime": 587.8211,
305
+ "eval_samples_per_second": 36.741,
306
+ "eval_steps_per_second": 4.593,
307
+ "eval_wer": 0.24392262503651294,
308
+ "step": 16000
309
+ },
310
+ {
311
+ "epoch": 4.39,
312
+ "learning_rate": 0.00017048489666136724,
313
+ "loss": 0.3416,
314
+ "step": 16800
315
+ },
316
+ {
317
+ "epoch": 4.39,
318
+ "eval_loss": 0.1961081475019455,
319
+ "eval_runtime": 586.8214,
320
+ "eval_samples_per_second": 36.803,
321
+ "eval_steps_per_second": 4.601,
322
+ "eval_wer": 0.24739541072993412,
323
+ "step": 16800
324
+ },
325
+ {
326
+ "epoch": 4.6,
327
+ "learning_rate": 0.0001641255961844197,
328
+ "loss": 0.3329,
329
+ "step": 17600
330
+ },
331
+ {
332
+ "epoch": 4.6,
333
+ "eval_loss": 0.19132652878761292,
334
+ "eval_runtime": 586.4075,
335
+ "eval_samples_per_second": 36.829,
336
+ "eval_steps_per_second": 4.604,
337
+ "eval_wer": 0.2405309791957418,
338
+ "step": 17600
339
+ }
340
+ ],
341
+ "max_steps": 38240,
342
+ "num_train_epochs": 10,
343
+ "total_flos": 5.592197758263177e+19,
344
+ "trial_name": null,
345
+ "trial_params": null
346
+ }
checkpoint-17600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5354e83b0770a05bb93186ef10aa8e3f89c954a3698919d22f8ad02ba33ef68f
3
+ size 3439