DewiBrynJones commited on
Commit
755f72a
1 Parent(s): 40370e3

End of training

Browse files
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - wer
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
17
 
18
- This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.4198
21
  - Wer: 0.3249
 
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
5
+ - automatic-speech-recognition
6
+ - DewiBrynJones/banc-trawsgrifiadau-bangor-normalized-with-ccv
7
  - generated_from_trainer
8
  metrics:
9
  - wer
 
17
 
18
  # wav2vec2-xlsr-53-ft-btb-ccv-cy
19
 
20
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-NORMALIZED-WITH-CCV - DEFAULT dataset.
21
  It achieves the following results on the evaluation set:
22
  - Loss: 0.4198
23
  - Wer: 0.3249
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.719665271966527,
3
+ "eval_loss": 0.41978269815444946,
4
+ "eval_runtime": 153.1864,
5
+ "eval_samples": 5656,
6
+ "eval_samples_per_second": 36.922,
7
+ "eval_steps_per_second": 4.615,
8
+ "eval_wer": 0.32485435958338016,
9
+ "total_flos": 1.0779565758702488e+19,
10
+ "train_loss": 1.1314451643136831,
11
+ "train_runtime": 7270.6329,
12
+ "train_samples": 30577,
13
+ "train_samples_per_second": 11.443,
14
+ "train_steps_per_second": 0.358
15
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.719665271966527,
3
+ "eval_loss": 0.41978269815444946,
4
+ "eval_runtime": 153.1864,
5
+ "eval_samples": 5656,
6
+ "eval_samples_per_second": 36.922,
7
+ "eval_steps_per_second": 4.615,
8
+ "eval_wer": 0.32485435958338016
9
+ }
runs/May23_14-31-09_7571a1a6d5fc/events.out.tfevents.1716478842.7571a1a6d5fc.291.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e24ea2692342cbe322bc19555697b2e1aadef8df98e7c3f8ad925fd1944325a0
3
+ size 406
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.719665271966527,
3
+ "total_flos": 1.0779565758702488e+19,
4
+ "train_loss": 1.1314451643136831,
5
+ "train_runtime": 7270.6329,
6
+ "train_samples": 30577,
7
+ "train_samples_per_second": 11.443,
8
+ "train_steps_per_second": 0.358
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,299 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.719665271966527,
5
+ "eval_steps": 100,
6
+ "global_step": 2600,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.10460251046025104,
13
+ "eval_loss": 3.505904197692871,
14
+ "eval_runtime": 155.3669,
15
+ "eval_samples_per_second": 36.404,
16
+ "eval_steps_per_second": 4.551,
17
+ "eval_wer": 1.0,
18
+ "step": 100
19
+ },
20
+ {
21
+ "epoch": 0.20920502092050208,
22
+ "eval_loss": 3.2686526775360107,
23
+ "eval_runtime": 151.6355,
24
+ "eval_samples_per_second": 37.3,
25
+ "eval_steps_per_second": 4.662,
26
+ "eval_wer": 1.0,
27
+ "step": 200
28
+ },
29
+ {
30
+ "epoch": 0.3138075313807531,
31
+ "eval_loss": 2.7369139194488525,
32
+ "eval_runtime": 151.6605,
33
+ "eval_samples_per_second": 37.294,
34
+ "eval_steps_per_second": 4.662,
35
+ "eval_wer": 1.0,
36
+ "step": 300
37
+ },
38
+ {
39
+ "epoch": 0.41841004184100417,
40
+ "eval_loss": 1.303921103477478,
41
+ "eval_runtime": 151.8516,
42
+ "eval_samples_per_second": 37.247,
43
+ "eval_steps_per_second": 4.656,
44
+ "eval_wer": 0.8544558745646836,
45
+ "step": 400
46
+ },
47
+ {
48
+ "epoch": 0.5230125523012552,
49
+ "grad_norm": 2.7574479579925537,
50
+ "learning_rate": 0.0002958,
51
+ "loss": 3.5426,
52
+ "step": 500
53
+ },
54
+ {
55
+ "epoch": 0.5230125523012552,
56
+ "eval_loss": 1.051817536354065,
57
+ "eval_runtime": 152.6837,
58
+ "eval_samples_per_second": 37.044,
59
+ "eval_steps_per_second": 4.63,
60
+ "eval_wer": 0.7738120075107124,
61
+ "step": 500
62
+ },
63
+ {
64
+ "epoch": 0.6276150627615062,
65
+ "eval_loss": 0.839580774307251,
66
+ "eval_runtime": 153.5079,
67
+ "eval_samples_per_second": 36.845,
68
+ "eval_steps_per_second": 4.606,
69
+ "eval_wer": 0.6148994559548073,
70
+ "step": 600
71
+ },
72
+ {
73
+ "epoch": 0.7322175732217573,
74
+ "eval_loss": 0.7548021674156189,
75
+ "eval_runtime": 152.4899,
76
+ "eval_samples_per_second": 37.091,
77
+ "eval_steps_per_second": 4.636,
78
+ "eval_wer": 0.5686636388438638,
79
+ "step": 700
80
+ },
81
+ {
82
+ "epoch": 0.8368200836820083,
83
+ "eval_loss": 0.6868751645088196,
84
+ "eval_runtime": 153.2403,
85
+ "eval_samples_per_second": 36.909,
86
+ "eval_steps_per_second": 4.614,
87
+ "eval_wer": 0.5185119802282101,
88
+ "step": 800
89
+ },
90
+ {
91
+ "epoch": 0.9414225941422594,
92
+ "eval_loss": 0.6501098871231079,
93
+ "eval_runtime": 152.6877,
94
+ "eval_samples_per_second": 37.043,
95
+ "eval_steps_per_second": 4.63,
96
+ "eval_wer": 0.4895283336810515,
97
+ "step": 900
98
+ },
99
+ {
100
+ "epoch": 1.0460251046025104,
101
+ "grad_norm": 1.2487365007400513,
102
+ "learning_rate": 0.00022957142857142856,
103
+ "loss": 0.7785,
104
+ "step": 1000
105
+ },
106
+ {
107
+ "epoch": 1.0460251046025104,
108
+ "eval_loss": 0.5817079544067383,
109
+ "eval_runtime": 153.4094,
110
+ "eval_samples_per_second": 36.869,
111
+ "eval_steps_per_second": 4.609,
112
+ "eval_wer": 0.4500649965495659,
113
+ "step": 1000
114
+ },
115
+ {
116
+ "epoch": 1.1506276150627615,
117
+ "eval_loss": 0.5646732449531555,
118
+ "eval_runtime": 152.9118,
119
+ "eval_samples_per_second": 36.989,
120
+ "eval_steps_per_second": 4.624,
121
+ "eval_wer": 0.4288006932965287,
122
+ "step": 1100
123
+ },
124
+ {
125
+ "epoch": 1.2552301255230125,
126
+ "eval_loss": 0.5424243807792664,
127
+ "eval_runtime": 152.6245,
128
+ "eval_samples_per_second": 37.058,
129
+ "eval_steps_per_second": 4.632,
130
+ "eval_wer": 0.42786987851262215,
131
+ "step": 1200
132
+ },
133
+ {
134
+ "epoch": 1.3598326359832635,
135
+ "eval_loss": 0.5264406800270081,
136
+ "eval_runtime": 153.2158,
137
+ "eval_samples_per_second": 36.915,
138
+ "eval_steps_per_second": 4.614,
139
+ "eval_wer": 0.4050809648376691,
140
+ "step": 1300
141
+ },
142
+ {
143
+ "epoch": 1.4644351464435146,
144
+ "eval_loss": 0.5098912715911865,
145
+ "eval_runtime": 153.9853,
146
+ "eval_samples_per_second": 36.731,
147
+ "eval_steps_per_second": 4.591,
148
+ "eval_wer": 0.3977467862817159,
149
+ "step": 1400
150
+ },
151
+ {
152
+ "epoch": 1.5690376569037658,
153
+ "grad_norm": 0.9644107222557068,
154
+ "learning_rate": 0.0001582857142857143,
155
+ "loss": 0.5795,
156
+ "step": 1500
157
+ },
158
+ {
159
+ "epoch": 1.5690376569037658,
160
+ "eval_loss": 0.5057937502861023,
161
+ "eval_runtime": 153.7283,
162
+ "eval_samples_per_second": 36.792,
163
+ "eval_steps_per_second": 4.599,
164
+ "eval_wer": 0.3952913610758935,
165
+ "step": 1500
166
+ },
167
+ {
168
+ "epoch": 1.6736401673640167,
169
+ "eval_loss": 0.48043006658554077,
170
+ "eval_runtime": 153.7864,
171
+ "eval_samples_per_second": 36.778,
172
+ "eval_steps_per_second": 4.597,
173
+ "eval_wer": 0.37892185970374415,
174
+ "step": 1600
175
+ },
176
+ {
177
+ "epoch": 1.778242677824268,
178
+ "eval_loss": 0.46723026037216187,
179
+ "eval_runtime": 153.9064,
180
+ "eval_samples_per_second": 36.75,
181
+ "eval_steps_per_second": 4.594,
182
+ "eval_wer": 0.3698062942337629,
183
+ "step": 1700
184
+ },
185
+ {
186
+ "epoch": 1.8828451882845187,
187
+ "eval_loss": 0.46052873134613037,
188
+ "eval_runtime": 153.9609,
189
+ "eval_samples_per_second": 36.737,
190
+ "eval_steps_per_second": 4.592,
191
+ "eval_wer": 0.3712185649403797,
192
+ "step": 1800
193
+ },
194
+ {
195
+ "epoch": 1.98744769874477,
196
+ "eval_loss": 0.4490562677383423,
197
+ "eval_runtime": 153.6441,
198
+ "eval_samples_per_second": 36.812,
199
+ "eval_steps_per_second": 4.602,
200
+ "eval_wer": 0.35561939304456675,
201
+ "step": 1900
202
+ },
203
+ {
204
+ "epoch": 2.092050209205021,
205
+ "grad_norm": 1.1409815549850464,
206
+ "learning_rate": 8.714285714285714e-05,
207
+ "loss": 0.5057,
208
+ "step": 2000
209
+ },
210
+ {
211
+ "epoch": 2.092050209205021,
212
+ "eval_loss": 0.44942042231559753,
213
+ "eval_runtime": 154.6428,
214
+ "eval_samples_per_second": 36.575,
215
+ "eval_steps_per_second": 4.572,
216
+ "eval_wer": 0.34534833336008086,
217
+ "step": 2000
218
+ },
219
+ {
220
+ "epoch": 2.196652719665272,
221
+ "eval_loss": 0.4431760907173157,
222
+ "eval_runtime": 154.0218,
223
+ "eval_samples_per_second": 36.722,
224
+ "eval_steps_per_second": 4.59,
225
+ "eval_wer": 0.3396832020028566,
226
+ "step": 2100
227
+ },
228
+ {
229
+ "epoch": 2.301255230125523,
230
+ "eval_loss": 0.43776625394821167,
231
+ "eval_runtime": 154.8082,
232
+ "eval_samples_per_second": 36.536,
233
+ "eval_steps_per_second": 4.567,
234
+ "eval_wer": 0.335061225144838,
235
+ "step": 2200
236
+ },
237
+ {
238
+ "epoch": 2.405857740585774,
239
+ "eval_loss": 0.42906874418258667,
240
+ "eval_runtime": 154.6416,
241
+ "eval_samples_per_second": 36.575,
242
+ "eval_steps_per_second": 4.572,
243
+ "eval_wer": 0.33104909245558567,
244
+ "step": 2300
245
+ },
246
+ {
247
+ "epoch": 2.510460251046025,
248
+ "eval_loss": 0.4279017746448517,
249
+ "eval_runtime": 154.5726,
250
+ "eval_samples_per_second": 36.591,
251
+ "eval_steps_per_second": 4.574,
252
+ "eval_wer": 0.3294281908491278,
253
+ "step": 2400
254
+ },
255
+ {
256
+ "epoch": 2.6150627615062763,
257
+ "grad_norm": 0.6568267345428467,
258
+ "learning_rate": 1.6e-05,
259
+ "loss": 0.3986,
260
+ "step": 2500
261
+ },
262
+ {
263
+ "epoch": 2.6150627615062763,
264
+ "eval_loss": 0.4233950078487396,
265
+ "eval_runtime": 154.8004,
266
+ "eval_samples_per_second": 36.537,
267
+ "eval_steps_per_second": 4.567,
268
+ "eval_wer": 0.3259456596748568,
269
+ "step": 2500
270
+ },
271
+ {
272
+ "epoch": 2.719665271966527,
273
+ "eval_loss": 0.41978269815444946,
274
+ "eval_runtime": 153.7416,
275
+ "eval_samples_per_second": 36.789,
276
+ "eval_steps_per_second": 4.599,
277
+ "eval_wer": 0.32485435958338016,
278
+ "step": 2600
279
+ },
280
+ {
281
+ "epoch": 2.719665271966527,
282
+ "step": 2600,
283
+ "total_flos": 1.0779565758702488e+19,
284
+ "train_loss": 1.1314451643136831,
285
+ "train_runtime": 7270.6329,
286
+ "train_samples_per_second": 11.443,
287
+ "train_steps_per_second": 0.358
288
+ }
289
+ ],
290
+ "logging_steps": 500,
291
+ "max_steps": 2600,
292
+ "num_input_tokens_seen": 0,
293
+ "num_train_epochs": 3,
294
+ "save_steps": 400,
295
+ "total_flos": 1.0779565758702488e+19,
296
+ "train_batch_size": 16,
297
+ "trial_name": null,
298
+ "trial_params": null
299
+ }