boumehdi commited on
Commit
a4fc8c1
1 Parent(s): 4b1d247

Upload 13 files

Browse files
config.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "C:\\workspace\\checkpoints\\",
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": 0.0,
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": 0.0,
58
+ "hidden_size": 1024,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 4096,
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_channel_length": 10,
64
+ "mask_channel_min_space": 1,
65
+ "mask_channel_other": 0.0,
66
+ "mask_channel_prob": 0.0,
67
+ "mask_channel_selection": "static",
68
+ "mask_feature_length": 10,
69
+ "mask_feature_min_masks": 0,
70
+ "mask_feature_prob": 0.0,
71
+ "mask_time_length": 10,
72
+ "mask_time_min_masks": 2,
73
+ "mask_time_min_space": 1,
74
+ "mask_time_other": 0.0,
75
+ "mask_time_prob": 0.05,
76
+ "mask_time_selection": "static",
77
+ "model_type": "wav2vec2",
78
+ "num_adapter_layers": 3,
79
+ "num_attention_heads": 16,
80
+ "num_codevector_groups": 2,
81
+ "num_codevectors_per_group": 320,
82
+ "num_conv_pos_embedding_groups": 16,
83
+ "num_conv_pos_embeddings": 128,
84
+ "num_feat_extract_layers": 7,
85
+ "num_hidden_layers": 24,
86
+ "num_negatives": 100,
87
+ "output_hidden_size": 1024,
88
+ "pad_token_id": 64,
89
+ "proj_codevector_dim": 768,
90
+ "tdnn_dilation": [
91
+ 1,
92
+ 2,
93
+ 3,
94
+ 1,
95
+ 1
96
+ ],
97
+ "tdnn_dim": [
98
+ 512,
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 1500
103
+ ],
104
+ "tdnn_kernel": [
105
+ 5,
106
+ 3,
107
+ 3,
108
+ 1,
109
+ 1
110
+ ],
111
+ "torch_dtype": "float32",
112
+ "transformers_version": "4.21.1",
113
+ "use_weighted_layer_sum": false,
114
+ "vocab_size": 65,
115
+ "xvector_output_dim": 512
116
+ }
flax_model.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b81a6443503273ade01e51ccf3ce7e8649ec7d8744626fb5fba5aac1da2cd396
3
+ size 1269577963
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e622da60d23e94e07cdb99ff234aaefa76f7c6fb21c57c6ab61ba5d19f8b9c7
3
+ size 2490594117
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
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4512cd385771733271cfd87d2c72759096c7a1a777295cc9b7fef89f555dfd2
3
+ size 1262168365
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:015a8d4a0507b7ea92a5d2b6e83f85108572344d34478a240dd0c974135c5e09
3
+ size 14575
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db575b4427150600ee37f6c6a1a316680eb05224eb75e119839f9b6fbad60117
3
+ size 557
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ff0fa787e174ecf06cbb98b01670eaefa203dc8ba7e6a8d2ed74e188444dfb3
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"}
trainer_state.json ADDED
@@ -0,0 +1,616 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 31.99304865938431,
5
+ "global_step": 4000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.79,
12
+ "learning_rate": 0.0001,
13
+ "loss": 0.0392,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 0.79,
18
+ "eval_loss": 0.23765751719474792,
19
+ "eval_runtime": 220.5891,
20
+ "eval_samples_per_second": 16.297,
21
+ "eval_steps_per_second": 2.04,
22
+ "eval_wer": 0.21887876816308827,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 1.6,
27
+ "learning_rate": 9.991993594875901e-05,
28
+ "loss": 0.0336,
29
+ "step": 200
30
+ },
31
+ {
32
+ "epoch": 1.6,
33
+ "eval_loss": 0.26165521144866943,
34
+ "eval_runtime": 157.9726,
35
+ "eval_samples_per_second": 22.757,
36
+ "eval_steps_per_second": 2.849,
37
+ "eval_wer": 0.21649316851008457,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 2.4,
42
+ "learning_rate": 9.983987189751803e-05,
43
+ "loss": 0.0293,
44
+ "step": 300
45
+ },
46
+ {
47
+ "epoch": 2.4,
48
+ "eval_loss": 0.28323182463645935,
49
+ "eval_runtime": 169.3448,
50
+ "eval_samples_per_second": 21.229,
51
+ "eval_steps_per_second": 2.657,
52
+ "eval_wer": 0.2197462589459987,
53
+ "step": 300
54
+ },
55
+ {
56
+ "epoch": 3.2,
57
+ "learning_rate": 9.975980784627703e-05,
58
+ "loss": 0.0283,
59
+ "step": 400
60
+ },
61
+ {
62
+ "epoch": 3.2,
63
+ "eval_loss": 0.29311421513557434,
64
+ "eval_runtime": 175.6178,
65
+ "eval_samples_per_second": 20.471,
66
+ "eval_steps_per_second": 2.562,
67
+ "eval_wer": 0.21508349598785512,
68
+ "step": 400
69
+ },
70
+ {
71
+ "epoch": 3.99,
72
+ "learning_rate": 9.967974379503603e-05,
73
+ "loss": 0.0274,
74
+ "step": 500
75
+ },
76
+ {
77
+ "epoch": 3.99,
78
+ "eval_loss": 0.30566948652267456,
79
+ "eval_runtime": 182.1945,
80
+ "eval_samples_per_second": 19.732,
81
+ "eval_steps_per_second": 2.47,
82
+ "eval_wer": 0.21828236824983735,
83
+ "step": 500
84
+ },
85
+ {
86
+ "epoch": 4.79,
87
+ "learning_rate": 9.959967974379504e-05,
88
+ "loss": 0.0264,
89
+ "step": 600
90
+ },
91
+ {
92
+ "epoch": 4.79,
93
+ "eval_loss": 0.30197781324386597,
94
+ "eval_runtime": 174.7969,
95
+ "eval_samples_per_second": 20.567,
96
+ "eval_steps_per_second": 2.574,
97
+ "eval_wer": 0.2171437865972674,
98
+ "step": 600
99
+ },
100
+ {
101
+ "epoch": 5.6,
102
+ "learning_rate": 9.951961569255405e-05,
103
+ "loss": 0.0259,
104
+ "step": 700
105
+ },
106
+ {
107
+ "epoch": 5.6,
108
+ "eval_loss": 0.3002428412437439,
109
+ "eval_runtime": 191.2838,
110
+ "eval_samples_per_second": 18.794,
111
+ "eval_steps_per_second": 2.353,
112
+ "eval_wer": 0.21725222294513122,
113
+ "step": 700
114
+ },
115
+ {
116
+ "epoch": 6.4,
117
+ "learning_rate": 9.943955164131305e-05,
118
+ "loss": 0.0254,
119
+ "step": 800
120
+ },
121
+ {
122
+ "epoch": 6.4,
123
+ "eval_loss": 0.3097267746925354,
124
+ "eval_runtime": 183.9687,
125
+ "eval_samples_per_second": 19.541,
126
+ "eval_steps_per_second": 2.446,
127
+ "eval_wer": 0.21752331381479073,
128
+ "step": 800
129
+ },
130
+ {
131
+ "epoch": 7.2,
132
+ "learning_rate": 9.936028823058447e-05,
133
+ "loss": 0.0252,
134
+ "step": 900
135
+ },
136
+ {
137
+ "epoch": 7.2,
138
+ "eval_loss": 0.2970833480358124,
139
+ "eval_runtime": 182.1563,
140
+ "eval_samples_per_second": 19.736,
141
+ "eval_steps_per_second": 2.47,
142
+ "eval_wer": 0.2170353502494036,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 7.99,
147
+ "learning_rate": 9.928022417934349e-05,
148
+ "loss": 0.0234,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 7.99,
153
+ "eval_loss": 0.31429246068000793,
154
+ "eval_runtime": 193.7999,
155
+ "eval_samples_per_second": 18.55,
156
+ "eval_steps_per_second": 2.322,
157
+ "eval_wer": 0.21405335068314899,
158
+ "step": 1000
159
+ },
160
+ {
161
+ "epoch": 8.79,
162
+ "learning_rate": 9.920016012810248e-05,
163
+ "loss": 0.0228,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 8.79,
168
+ "eval_loss": 0.32985326647758484,
169
+ "eval_runtime": 194.3544,
170
+ "eval_samples_per_second": 18.497,
171
+ "eval_steps_per_second": 2.315,
172
+ "eval_wer": 0.2142702233788766,
173
+ "step": 1100
174
+ },
175
+ {
176
+ "epoch": 9.6,
177
+ "learning_rate": 9.912009607686149e-05,
178
+ "loss": 0.0236,
179
+ "step": 1200
180
+ },
181
+ {
182
+ "epoch": 9.6,
183
+ "eval_loss": 0.3165593147277832,
184
+ "eval_runtime": 185.0001,
185
+ "eval_samples_per_second": 19.432,
186
+ "eval_steps_per_second": 2.432,
187
+ "eval_wer": 0.21833658642376924,
188
+ "step": 1200
189
+ },
190
+ {
191
+ "epoch": 10.4,
192
+ "learning_rate": 9.90400320256205e-05,
193
+ "loss": 0.0241,
194
+ "step": 1300
195
+ },
196
+ {
197
+ "epoch": 10.4,
198
+ "eval_loss": 0.3285478949546814,
199
+ "eval_runtime": 181.9532,
200
+ "eval_samples_per_second": 19.758,
201
+ "eval_steps_per_second": 2.473,
202
+ "eval_wer": 0.2192582953806116,
203
+ "step": 1300
204
+ },
205
+ {
206
+ "epoch": 11.2,
207
+ "learning_rate": 9.895996797437951e-05,
208
+ "loss": 0.0243,
209
+ "step": 1400
210
+ },
211
+ {
212
+ "epoch": 11.2,
213
+ "eval_loss": 0.3187803626060486,
214
+ "eval_runtime": 200.4409,
215
+ "eval_samples_per_second": 17.935,
216
+ "eval_steps_per_second": 2.245,
217
+ "eval_wer": 0.22099327694643245,
218
+ "step": 1400
219
+ },
220
+ {
221
+ "epoch": 11.99,
222
+ "learning_rate": 9.887990392313852e-05,
223
+ "loss": 0.026,
224
+ "step": 1500
225
+ },
226
+ {
227
+ "epoch": 11.99,
228
+ "eval_loss": 0.32988375425338745,
229
+ "eval_runtime": 199.5224,
230
+ "eval_samples_per_second": 18.018,
231
+ "eval_steps_per_second": 2.255,
232
+ "eval_wer": 0.22375840381695944,
233
+ "step": 1500
234
+ },
235
+ {
236
+ "epoch": 12.79,
237
+ "learning_rate": 9.879983987189752e-05,
238
+ "loss": 0.0259,
239
+ "step": 1600
240
+ },
241
+ {
242
+ "epoch": 12.79,
243
+ "eval_loss": 0.3099309206008911,
244
+ "eval_runtime": 198.0592,
245
+ "eval_samples_per_second": 18.151,
246
+ "eval_steps_per_second": 2.272,
247
+ "eval_wer": 0.22045109520711342,
248
+ "step": 1600
249
+ },
250
+ {
251
+ "epoch": 13.6,
252
+ "learning_rate": 9.871977582065654e-05,
253
+ "loss": 0.0255,
254
+ "step": 1700
255
+ },
256
+ {
257
+ "epoch": 13.6,
258
+ "eval_loss": 0.30543622374534607,
259
+ "eval_runtime": 202.823,
260
+ "eval_samples_per_second": 17.725,
261
+ "eval_steps_per_second": 2.219,
262
+ "eval_wer": 0.22402949468661895,
263
+ "step": 1700
264
+ },
265
+ {
266
+ "epoch": 14.4,
267
+ "learning_rate": 9.863971176941553e-05,
268
+ "loss": 0.0253,
269
+ "step": 1800
270
+ },
271
+ {
272
+ "epoch": 14.4,
273
+ "eval_loss": 0.31324318051338196,
274
+ "eval_runtime": 199.664,
275
+ "eval_samples_per_second": 18.005,
276
+ "eval_steps_per_second": 2.254,
277
+ "eval_wer": 0.21779440468445022,
278
+ "step": 1800
279
+ },
280
+ {
281
+ "epoch": 15.2,
282
+ "learning_rate": 9.855964771817454e-05,
283
+ "loss": 0.0244,
284
+ "step": 1900
285
+ },
286
+ {
287
+ "epoch": 15.2,
288
+ "eval_loss": 0.33187857270240784,
289
+ "eval_runtime": 197.0859,
290
+ "eval_samples_per_second": 18.241,
291
+ "eval_steps_per_second": 2.283,
292
+ "eval_wer": 0.22121014964216004,
293
+ "step": 1900
294
+ },
295
+ {
296
+ "epoch": 15.99,
297
+ "learning_rate": 9.847958366693355e-05,
298
+ "loss": 0.0231,
299
+ "step": 2000
300
+ },
301
+ {
302
+ "epoch": 15.99,
303
+ "eval_loss": 0.33831512928009033,
304
+ "eval_runtime": 200.252,
305
+ "eval_samples_per_second": 17.952,
306
+ "eval_steps_per_second": 2.247,
307
+ "eval_wer": 0.21920407720667967,
308
+ "step": 2000
309
+ },
310
+ {
311
+ "epoch": 16.79,
312
+ "learning_rate": 9.839951961569256e-05,
313
+ "loss": 0.0235,
314
+ "step": 2100
315
+ },
316
+ {
317
+ "epoch": 16.79,
318
+ "eval_loss": 0.31389620900154114,
319
+ "eval_runtime": 198.2191,
320
+ "eval_samples_per_second": 18.136,
321
+ "eval_steps_per_second": 2.27,
322
+ "eval_wer": 0.21622207764042506,
323
+ "step": 2100
324
+ },
325
+ {
326
+ "epoch": 17.6,
327
+ "learning_rate": 9.831945556445156e-05,
328
+ "loss": 0.0227,
329
+ "step": 2200
330
+ },
331
+ {
332
+ "epoch": 17.6,
333
+ "eval_loss": 0.32037118077278137,
334
+ "eval_runtime": 202.9198,
335
+ "eval_samples_per_second": 17.716,
336
+ "eval_steps_per_second": 2.218,
337
+ "eval_wer": 0.217360659292995,
338
+ "step": 2200
339
+ },
340
+ {
341
+ "epoch": 18.4,
342
+ "learning_rate": 9.823939151321058e-05,
343
+ "loss": 0.0228,
344
+ "step": 2300
345
+ },
346
+ {
347
+ "epoch": 18.4,
348
+ "eval_loss": 0.32169201970100403,
349
+ "eval_runtime": 193.9,
350
+ "eval_samples_per_second": 18.54,
351
+ "eval_steps_per_second": 2.321,
352
+ "eval_wer": 0.21757753198872262,
353
+ "step": 2300
354
+ },
355
+ {
356
+ "epoch": 19.2,
357
+ "learning_rate": 9.815932746196959e-05,
358
+ "loss": 0.0217,
359
+ "step": 2400
360
+ },
361
+ {
362
+ "epoch": 19.2,
363
+ "eval_loss": 0.3112569749355316,
364
+ "eval_runtime": 198.2135,
365
+ "eval_samples_per_second": 18.137,
366
+ "eval_steps_per_second": 2.27,
367
+ "eval_wer": 0.2170895684233355,
368
+ "step": 2400
369
+ },
370
+ {
371
+ "epoch": 19.99,
372
+ "learning_rate": 9.807926341072858e-05,
373
+ "loss": 0.0212,
374
+ "step": 2500
375
+ },
376
+ {
377
+ "epoch": 19.99,
378
+ "eval_loss": 0.31596991419792175,
379
+ "eval_runtime": 199.2987,
380
+ "eval_samples_per_second": 18.038,
381
+ "eval_steps_per_second": 2.258,
382
+ "eval_wer": 0.21351116894382996,
383
+ "step": 2500
384
+ },
385
+ {
386
+ "epoch": 20.79,
387
+ "learning_rate": 9.79991993594876e-05,
388
+ "loss": 0.0216,
389
+ "step": 2600
390
+ },
391
+ {
392
+ "epoch": 20.79,
393
+ "eval_loss": 0.3226161003112793,
394
+ "eval_runtime": 199.747,
395
+ "eval_samples_per_second": 17.998,
396
+ "eval_steps_per_second": 2.253,
397
+ "eval_wer": 0.21378225981348947,
398
+ "step": 2600
399
+ },
400
+ {
401
+ "epoch": 21.6,
402
+ "learning_rate": 9.79191353082466e-05,
403
+ "loss": 0.0242,
404
+ "step": 2700
405
+ },
406
+ {
407
+ "epoch": 21.6,
408
+ "eval_loss": 0.3281223177909851,
409
+ "eval_runtime": 197.1976,
410
+ "eval_samples_per_second": 18.23,
411
+ "eval_steps_per_second": 2.282,
412
+ "eval_wer": 0.2169269139015398,
413
+ "step": 2700
414
+ },
415
+ {
416
+ "epoch": 22.4,
417
+ "learning_rate": 9.783907125700561e-05,
418
+ "loss": 0.0245,
419
+ "step": 2800
420
+ },
421
+ {
422
+ "epoch": 22.4,
423
+ "eval_loss": 0.3078162968158722,
424
+ "eval_runtime": 198.6659,
425
+ "eval_samples_per_second": 18.096,
426
+ "eval_steps_per_second": 2.265,
427
+ "eval_wer": 0.2176317501626545,
428
+ "step": 2800
429
+ },
430
+ {
431
+ "epoch": 23.2,
432
+ "learning_rate": 9.775900720576461e-05,
433
+ "loss": 0.0229,
434
+ "step": 2900
435
+ },
436
+ {
437
+ "epoch": 23.2,
438
+ "eval_loss": 0.31995320320129395,
439
+ "eval_runtime": 204.328,
440
+ "eval_samples_per_second": 17.594,
441
+ "eval_steps_per_second": 2.202,
442
+ "eval_wer": 0.22045109520711342,
443
+ "step": 2900
444
+ },
445
+ {
446
+ "epoch": 23.99,
447
+ "learning_rate": 9.767894315452363e-05,
448
+ "loss": 0.0226,
449
+ "step": 3000
450
+ },
451
+ {
452
+ "epoch": 23.99,
453
+ "eval_loss": 0.33075791597366333,
454
+ "eval_runtime": 205.0698,
455
+ "eval_samples_per_second": 17.531,
456
+ "eval_steps_per_second": 2.194,
457
+ "eval_wer": 0.2171437865972674,
458
+ "step": 3000
459
+ },
460
+ {
461
+ "epoch": 24.79,
462
+ "learning_rate": 9.759887910328262e-05,
463
+ "loss": 0.0213,
464
+ "step": 3100
465
+ },
466
+ {
467
+ "epoch": 24.79,
468
+ "eval_loss": 0.3419627845287323,
469
+ "eval_runtime": 206.2312,
470
+ "eval_samples_per_second": 17.432,
471
+ "eval_steps_per_second": 2.182,
472
+ "eval_wer": 0.21936673172847537,
473
+ "step": 3100
474
+ },
475
+ {
476
+ "epoch": 25.6,
477
+ "learning_rate": 9.751881505204163e-05,
478
+ "loss": 0.0212,
479
+ "step": 3200
480
+ },
481
+ {
482
+ "epoch": 25.6,
483
+ "eval_loss": 0.3389272093772888,
484
+ "eval_runtime": 205.5665,
485
+ "eval_samples_per_second": 17.488,
486
+ "eval_steps_per_second": 2.189,
487
+ "eval_wer": 0.21519193233571893,
488
+ "step": 3200
489
+ },
490
+ {
491
+ "epoch": 26.4,
492
+ "learning_rate": 9.743875100080065e-05,
493
+ "loss": 0.0208,
494
+ "step": 3300
495
+ },
496
+ {
497
+ "epoch": 26.4,
498
+ "eval_loss": 0.32781022787094116,
499
+ "eval_runtime": 208.0862,
500
+ "eval_samples_per_second": 17.276,
501
+ "eval_steps_per_second": 2.163,
502
+ "eval_wer": 0.21275211450878334,
503
+ "step": 3300
504
+ },
505
+ {
506
+ "epoch": 27.2,
507
+ "learning_rate": 9.735868694955965e-05,
508
+ "loss": 0.0212,
509
+ "step": 3400
510
+ },
511
+ {
512
+ "epoch": 27.2,
513
+ "eval_loss": 0.3169388771057129,
514
+ "eval_runtime": 206.029,
515
+ "eval_samples_per_second": 17.449,
516
+ "eval_steps_per_second": 2.184,
517
+ "eval_wer": 0.21497505963999133,
518
+ "step": 3400
519
+ },
520
+ {
521
+ "epoch": 27.99,
522
+ "learning_rate": 9.727862289831866e-05,
523
+ "loss": 0.0213,
524
+ "step": 3500
525
+ },
526
+ {
527
+ "epoch": 27.99,
528
+ "eval_loss": 0.31786179542541504,
529
+ "eval_runtime": 198.632,
530
+ "eval_samples_per_second": 18.099,
531
+ "eval_steps_per_second": 2.265,
532
+ "eval_wer": 0.21757753198872262,
533
+ "step": 3500
534
+ },
535
+ {
536
+ "epoch": 28.79,
537
+ "learning_rate": 9.719855884707766e-05,
538
+ "loss": 0.0213,
539
+ "step": 3600
540
+ },
541
+ {
542
+ "epoch": 28.79,
543
+ "eval_loss": 0.32053136825561523,
544
+ "eval_runtime": 201.1109,
545
+ "eval_samples_per_second": 17.876,
546
+ "eval_steps_per_second": 2.238,
547
+ "eval_wer": 0.2203426588592496,
548
+ "step": 3600
549
+ },
550
+ {
551
+ "epoch": 29.6,
552
+ "learning_rate": 9.711849479583668e-05,
553
+ "loss": 0.0217,
554
+ "step": 3700
555
+ },
556
+ {
557
+ "epoch": 29.6,
558
+ "eval_loss": 0.33174052834510803,
559
+ "eval_runtime": 191.0157,
560
+ "eval_samples_per_second": 18.82,
561
+ "eval_steps_per_second": 2.356,
562
+ "eval_wer": 0.2168726957276079,
563
+ "step": 3700
564
+ },
565
+ {
566
+ "epoch": 30.4,
567
+ "learning_rate": 9.703843074459567e-05,
568
+ "loss": 0.0221,
569
+ "step": 3800
570
+ },
571
+ {
572
+ "epoch": 30.4,
573
+ "eval_loss": 0.3100583255290985,
574
+ "eval_runtime": 190.6873,
575
+ "eval_samples_per_second": 18.853,
576
+ "eval_steps_per_second": 2.36,
577
+ "eval_wer": 0.21774018651051832,
578
+ "step": 3800
579
+ },
580
+ {
581
+ "epoch": 31.2,
582
+ "learning_rate": 9.695836669335468e-05,
583
+ "loss": 0.021,
584
+ "step": 3900
585
+ },
586
+ {
587
+ "epoch": 31.2,
588
+ "eval_loss": 0.33132240176200867,
589
+ "eval_runtime": 187.5625,
590
+ "eval_samples_per_second": 19.167,
591
+ "eval_steps_per_second": 2.399,
592
+ "eval_wer": 0.21741487746692692,
593
+ "step": 3900
594
+ },
595
+ {
596
+ "epoch": 31.99,
597
+ "learning_rate": 9.68783026421137e-05,
598
+ "loss": 0.0212,
599
+ "step": 4000
600
+ },
601
+ {
602
+ "epoch": 31.99,
603
+ "eval_loss": 0.32063281536102295,
604
+ "eval_runtime": 192.7124,
605
+ "eval_samples_per_second": 18.655,
606
+ "eval_steps_per_second": 2.335,
607
+ "eval_wer": 0.21340273259596618,
608
+ "step": 4000
609
+ }
610
+ ],
611
+ "max_steps": 125000,
612
+ "num_train_epochs": 1000,
613
+ "total_flos": 9.112861160368682e+19,
614
+ "trial_name": null,
615
+ "trial_params": null
616
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4775b7d54e5ed7fc71d018fd7cb01140ef925d4333446c094a86299d55c0701
3
+ size 3323
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"ا": 0, "_": 1, "ك": 3, "ن": 4, "@": 5, "ذ": 6, "ٖ": 7, "ش": 8, "غ": 9, "آ": 10, "ص": 11, "ي": 12, "ث": 13, "n": 14, "ج": 15, "d": 16, "ئ": 17, "ة": 18, "ه": 19, "ز": 20, "ع": 21, "ف": 22, "i": 23, "r": 24, "v": 25, "ڸ": 26, "k": 27, "ِ": 28, "ء": 29, "ر": 30, "s": 31, "t": 32, "ى": 33, "ـ": 34, " ": 35, "ؤ": 36, "ق": 37, "ض": 38, "م": 39, "ل": 40, "…": 41, "ط": 42, "ت": 43, "ّ": 44, "c": 45, "ظ": 46, "e": 47, "؟": 48, "h": 49, "ب": 50, "o": 51, "س": 52, "a": 53, "د": 54, "p": 55, "أ": 56, "ْ": 57, "\n": 58, "خ": 59, "ح": 60, "و": 61, "إ": 62, "|": 2, "[UNK]": 63, "[PAD]": 64}