Hugging Face's logo

ihanif
/
whisper_small_ps_augmented

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Pashto
whisper
whisper-event
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
ihanif commited on
Commit
0e2b556
•
1 Parent(s): e8768b1

add checkpoint 300

Browse files
Files changed (9) hide show
  1. checkpoint-300/config.json +42 -0
  2. checkpoint-300/optimizer.pt +3 -0
  3. checkpoint-300/preprocessor_config.json +0 -0
  4. checkpoint-300/pytorch_model.bin +3 -0
  5. checkpoint-300/rng_state.pth +3 -0
  6. checkpoint-300/scaler.pt +3 -0
  7. checkpoint-300/scheduler.pt +3 -0
  8. checkpoint-300/trainer_state.json +226 -0
  9. checkpoint-300/training_args.bin +3 -0
checkpoint-300/config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-small",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50257
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "d_model": 768,
15
+ "decoder_attention_heads": 12,
16
+ "decoder_ffn_dim": 3072,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 12,
22
+ "encoder_ffn_dim": 3072,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 12,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": null,
27
+ "init_std": 0.02,
28
+ "is_encoder_decoder": true,
29
+ "max_length": 448,
30
+ "max_source_positions": 1500,
31
+ "max_target_positions": 448,
32
+ "model_type": "whisper",
33
+ "num_hidden_layers": 12,
34
+ "num_mel_bins": 80,
35
+ "pad_token_id": 50257,
36
+ "scale_embedding": false,
37
+ "suppress_tokens": [],
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.26.0.dev0",
40
+ "use_cache": false,
41
+ "vocab_size": 51865
42
+ }
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a29af7e42a8cc291c7dc41a87ab9bec819ca5e7407672e72ed2528611450ee9
3
+ size 1934161093
checkpoint-300/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3659cbd57caabfa6834081314e1044f720d4f82db5a36a341158bdc9fc0cf4f2
3
+ size 967102601
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0968b0e39b19cec9553e062832c01be226fad3fe583411f84809dfd1a9c08ee1
3
+ size 14511
checkpoint-300/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ef418a6ac4669ffd3c7cffe1d7df12a1bb116658f99f16e613cb28c6cbef5cf
3
+ size 557
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6c1d870d593d2993fc02f7df3f7be41e6fce04eb6216560256b783d35551c95
3
+ size 627
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,226 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 53.62439467312349,
3
+ "best_model_checkpoint": "./checkpoint-300",
4
+ "epoch": 3.566371681415929,
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": 0.12,
12
+ "learning_rate": 3e-06,
13
+ "loss": 4.1495,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.24,
18
+ "learning_rate": 6.333333333333333e-06,
19
+ "loss": 2.9287,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.35,
24
+ "learning_rate": 9.666666666666667e-06,
25
+ "loss": 2.0462,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.47,
30
+ "learning_rate": 9.666666666666667e-06,
31
+ "loss": 1.6138,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.59,
36
+ "learning_rate": 9.296296296296296e-06,
37
+ "loss": 1.3862,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.71,
42
+ "learning_rate": 8.925925925925927e-06,
43
+ "loss": 1.2604,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.83,
48
+ "learning_rate": 8.555555555555556e-06,
49
+ "loss": 1.1436,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.94,
54
+ "learning_rate": 8.185185185185187e-06,
55
+ "loss": 1.168,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 1.07,
60
+ "learning_rate": 7.814814814814816e-06,
61
+ "loss": 1.1041,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 1.19,
66
+ "learning_rate": 7.444444444444445e-06,
67
+ "loss": 0.9683,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 1.19,
72
+ "eval_cer": 131.61659035460045,
73
+ "eval_loss": 0.8811978697776794,
74
+ "eval_runtime": 593.3196,
75
+ "eval_samples_per_second": 0.863,
76
+ "eval_steps_per_second": 0.431,
77
+ "eval_wer": 139.37651331719127,
78
+ "step": 100
79
+ },
80
+ {
81
+ "epoch": 1.31,
82
+ "learning_rate": 7.074074074074074e-06,
83
+ "loss": 0.909,
84
+ "step": 110
85
+ },
86
+ {
87
+ "epoch": 1.42,
88
+ "learning_rate": 6.703703703703704e-06,
89
+ "loss": 0.9213,
90
+ "step": 120
91
+ },
92
+ {
93
+ "epoch": 1.54,
94
+ "learning_rate": 6.333333333333333e-06,
95
+ "loss": 0.9092,
96
+ "step": 130
97
+ },
98
+ {
99
+ "epoch": 1.66,
100
+ "learning_rate": 5.962962962962963e-06,
101
+ "loss": 0.8481,
102
+ "step": 140
103
+ },
104
+ {
105
+ "epoch": 1.78,
106
+ "learning_rate": 5.5925925925925926e-06,
107
+ "loss": 0.8471,
108
+ "step": 150
109
+ },
110
+ {
111
+ "epoch": 1.9,
112
+ "learning_rate": 5.2222222222222226e-06,
113
+ "loss": 0.8504,
114
+ "step": 160
115
+ },
116
+ {
117
+ "epoch": 2.02,
118
+ "learning_rate": 4.851851851851852e-06,
119
+ "loss": 0.8264,
120
+ "step": 170
121
+ },
122
+ {
123
+ "epoch": 2.14,
124
+ "learning_rate": 4.481481481481482e-06,
125
+ "loss": 0.7236,
126
+ "step": 180
127
+ },
128
+ {
129
+ "epoch": 2.26,
130
+ "learning_rate": 4.111111111111111e-06,
131
+ "loss": 0.6898,
132
+ "step": 190
133
+ },
134
+ {
135
+ "epoch": 2.38,
136
+ "learning_rate": 3.740740740740741e-06,
137
+ "loss": 0.6848,
138
+ "step": 200
139
+ },
140
+ {
141
+ "epoch": 2.38,
142
+ "eval_cer": 151.33685371478225,
143
+ "eval_loss": 0.7542899250984192,
144
+ "eval_runtime": 551.6472,
145
+ "eval_samples_per_second": 0.928,
146
+ "eval_steps_per_second": 0.464,
147
+ "eval_wer": 145.9972760290557,
148
+ "step": 200
149
+ },
150
+ {
151
+ "epoch": 2.5,
152
+ "learning_rate": 3.3703703703703705e-06,
153
+ "loss": 0.7021,
154
+ "step": 210
155
+ },
156
+ {
157
+ "epoch": 2.61,
158
+ "learning_rate": 3e-06,
159
+ "loss": 0.6956,
160
+ "step": 220
161
+ },
162
+ {
163
+ "epoch": 2.73,
164
+ "learning_rate": 2.6296296296296297e-06,
165
+ "loss": 0.629,
166
+ "step": 230
167
+ },
168
+ {
169
+ "epoch": 2.85,
170
+ "learning_rate": 2.2592592592592592e-06,
171
+ "loss": 0.6661,
172
+ "step": 240
173
+ },
174
+ {
175
+ "epoch": 2.97,
176
+ "learning_rate": 1.888888888888889e-06,
177
+ "loss": 0.6251,
178
+ "step": 250
179
+ },
180
+ {
181
+ "epoch": 3.09,
182
+ "learning_rate": 1.5185185185185186e-06,
183
+ "loss": 0.6852,
184
+ "step": 260
185
+ },
186
+ {
187
+ "epoch": 3.21,
188
+ "learning_rate": 1.1481481481481482e-06,
189
+ "loss": 0.5772,
190
+ "step": 270
191
+ },
192
+ {
193
+ "epoch": 3.33,
194
+ "learning_rate": 7.777777777777779e-07,
195
+ "loss": 0.5592,
196
+ "step": 280
197
+ },
198
+ {
199
+ "epoch": 3.45,
200
+ "learning_rate": 4.074074074074075e-07,
201
+ "loss": 0.5845,
202
+ "step": 290
203
+ },
204
+ {
205
+ "epoch": 3.57,
206
+ "learning_rate": 3.703703703703704e-08,
207
+ "loss": 0.5548,
208
+ "step": 300
209
+ },
210
+ {
211
+ "epoch": 3.57,
212
+ "eval_cer": 22.68473647271147,
213
+ "eval_loss": 0.6978507041931152,
214
+ "eval_runtime": 832.954,
215
+ "eval_samples_per_second": 0.615,
216
+ "eval_steps_per_second": 0.307,
217
+ "eval_wer": 53.62439467312349,
218
+ "step": 300
219
+ }
220
+ ],
221
+ "max_steps": 300,
222
+ "num_train_epochs": 4,
223
+ "total_flos": 2.79033223569408e+18,
224
+ "trial_name": null,
225
+ "trial_params": null
226
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcf3a24325bb49c25270193816ed6035a253ca3ae300c31ffcec0afeb4229266
3
+ size 3579