Hugging Face's logo

A-POR-LOS-8000
/
distilhubert-finetuned-cry-detector

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Marcos12886 commited on
Commit
eb22acc
•
1 Parent(s): 9487b76

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +25 -25
  2. checkpoint-500/config.json +77 -0
  3. checkpoint-500/model.safetensors +3 -0
  4. checkpoint-500/optimizer.pt +3 -0
  5. checkpoint-500/rng_state.pth +3 -0
  6. checkpoint-500/scheduler.pt +3 -0
  7. checkpoint-500/trainer_state.json +222 -0
  8. checkpoint-500/training_args.bin +3 -0
  9. checkpoint-600/config.json +77 -0
  10. checkpoint-600/model.safetensors +3 -0
  11. checkpoint-600/optimizer.pt +3 -0
  12. checkpoint-600/rng_state.pth +3 -0
  13. checkpoint-600/scheduler.pt +3 -0
  14. checkpoint-600/trainer_state.json +258 -0
  15. checkpoint-600/training_args.bin +3 -0
  16. checkpoint-700/config.json +77 -0
  17. checkpoint-700/model.safetensors +3 -0
  18. checkpoint-700/optimizer.pt +3 -0
  19. checkpoint-700/rng_state.pth +3 -0
  20. checkpoint-700/scheduler.pt +3 -0
  21. checkpoint-700/trainer_state.json +294 -0
  22. checkpoint-700/training_args.bin +3 -0
  23. checkpoint-800/config.json +77 -0
  24. checkpoint-800/model.safetensors +3 -0
  25. checkpoint-800/optimizer.pt +3 -0
  26. checkpoint-800/rng_state.pth +3 -0
  27. checkpoint-800/scheduler.pt +3 -0
  28. checkpoint-800/trainer_state.json +330 -0
  29. checkpoint-800/training_args.bin +3 -0
  30. checkpoint-860/config.json +77 -0
  31. checkpoint-860/model.safetensors +3 -0
  32. checkpoint-860/optimizer.pt +3 -0
  33. checkpoint-860/rng_state.pth +3 -0
  34. checkpoint-860/scheduler.pt +3 -0
  35. checkpoint-860/trainer_state.json +337 -0
  36. checkpoint-860/training_args.bin +3 -0
  37. model.safetensors +1 -1
  38. runs/Sep11_15-20-39_ubumarcos/events.out.tfevents.1726060840.ubumarcos +3 -0
  39. runs/Sep11_15-21-14_ubumarcos/events.out.tfevents.1726060876.ubumarcos +3 -0
  40. runs/Sep11_15-25-59_ubumarcos/events.out.tfevents.1726061160.ubumarcos +3 -0
  41. runs/Sep11_15-25-59_ubumarcos/events.out.tfevents.1726061977.ubumarcos +3 -0
  42. runs/Sep11_15-42-12_ubumarcos/events.out.tfevents.1726062133.ubumarcos +3 -0
  43. runs/Sep11_15-50-04_ubumarcos/events.out.tfevents.1726062605.ubumarcos +3 -0
  44. runs/Sep11_15-50-31_ubumarcos/events.out.tfevents.1726062632.ubumarcos +3 -0
  45. runs/Sep11_15-50-57_ubumarcos/events.out.tfevents.1726062658.ubumarcos +3 -0
  46. runs/Sep11_15-51-52_ubumarcos/events.out.tfevents.1726062713.ubumarcos +3 -0
  47. runs/Sep11_15-52-20_ubumarcos/events.out.tfevents.1726062741.ubumarcos +3 -0
  48. runs/Sep11_15-52-43_ubumarcos/events.out.tfevents.1726062764.ubumarcos +3 -0
  49. runs/Sep11_15-54-06_ubumarcos/events.out.tfevents.1726062847.ubumarcos +3 -0
  50. runs/Sep11_15-54-19_ubumarcos/events.out.tfevents.1726062860.ubumarcos +3 -0
README.md CHANGED
@@ -6,9 +6,9 @@ tags:
6
  - generated_from_trainer
7
  metrics:
8
  - accuracy
 
9
  - precision
10
  - recall
11
- - f1
12
  model-index:
13
  - name: distilhubert-finetuned-cry-detector
14
  results: []
@@ -21,11 +21,12 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.0338
25
- - Accuracy: 0.9912
26
- - Precision: 0.9912
27
- - Recall: 0.9912
28
- - F1: 0.9912
 
29
 
30
  ## Model description
31
 
@@ -45,30 +46,29 @@ More information needed
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 3e-05
48
- - train_batch_size: 8
49
- - eval_batch_size: 8
50
  - seed: 123
51
- - gradient_accumulation_steps: 8
52
- - total_train_batch_size: 64
53
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
54
- - lr_scheduler_type: cosine
55
- - lr_scheduler_warmup_ratio: 0.001
56
- - num_epochs: 10
 
57
 
58
  ### Training results
59
 
60
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
61
- |:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
62
- | No log | 0.9956 | 85 | 0.1197 | 0.9538 | 0.9535 | 0.9538 | 0.9536 |
63
- | No log | 1.9912 | 170 | 0.0601 | 0.9832 | 0.9831 | 0.9832 | 0.9831 |
64
- | No log | 2.9985 | 256 | 0.0506 | 0.9868 | 0.9868 | 0.9868 | 0.9868 |
65
- | No log | 3.9941 | 341 | 0.0461 | 0.9861 | 0.9864 | 0.9861 | 0.9862 |
66
- | No log | 4.9898 | 426 | 0.0439 | 0.9890 | 0.9891 | 0.9890 | 0.9890 |
67
- | 0.0779 | 5.9971 | 512 | 0.0396 | 0.9905 | 0.9905 | 0.9905 | 0.9905 |
68
- | 0.0779 | 6.9927 | 597 | 0.0350 | 0.9919 | 0.9919 | 0.9919 | 0.9919 |
69
- | 0.0779 | 8.0 | 683 | 0.0335 | 0.9912 | 0.9912 | 0.9912 | 0.9912 |
70
- | 0.0779 | 8.9956 | 768 | 0.0337 | 0.9912 | 0.9912 | 0.9912 | 0.9912 |
71
- | 0.0779 | 9.9561 | 850 | 0.0338 | 0.9912 | 0.9912 | 0.9912 | 0.9912 |
72
 
73
 
74
  ### Framework versions
 
6
  - generated_from_trainer
7
  metrics:
8
  - accuracy
9
+ - f1
10
  - precision
11
  - recall
 
12
  model-index:
13
  - name: distilhubert-finetuned-cry-detector
14
  results: []
 
21
 
22
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.2255
25
+ - Accuracy: 0.9883
26
+ - F1: 0.9883
27
+ - Precision: 0.9883
28
+ - Recall: 0.9883
29
+ - Confusion Matrix: [[960, 10], [6, 389]]
30
 
31
  ## Model description
32
 
 
46
 
47
  The following hyperparameters were used during training:
48
  - learning_rate: 3e-05
49
+ - train_batch_size: 64
50
+ - eval_batch_size: 64
51
  - seed: 123
52
+ - gradient_accumulation_steps: 2
53
+ - total_train_batch_size: 128
54
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
55
+ - lr_scheduler_type: cosine_with_restarts
56
+ - lr_scheduler_warmup_ratio: 0.1
57
+ - num_epochs: 20
58
+ - label_smoothing_factor: 0.1
59
 
60
  ### Training results
61
 
62
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | Confusion Matrix |
63
+ |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:----------------------:|
64
+ | 0.3124 | 2.3256 | 100 | 0.2739 | 0.9641 | 0.9640 | 0.9640 | 0.9641 | [[948, 22], [27, 368]] |
65
+ | 0.2337 | 4.6512 | 200 | 0.2385 | 0.9736 | 0.9737 | 0.9737 | 0.9736 | [[950, 20], [16, 379]] |
66
+ | 0.2064 | 6.9767 | 300 | 0.2295 | 0.9832 | 0.9832 | 0.9832 | 0.9832 | [[958, 12], [11, 384]] |
67
+ | 0.2023 | 9.3023 | 400 | 0.2277 | 0.9868 | 0.9869 | 0.9870 | 0.9868 | [[957, 13], [5, 390]] |
68
+ | 0.2003 | 11.6279 | 500 | 0.2254 | 0.9875 | 0.9876 | 0.9876 | 0.9875 | [[960, 10], [7, 388]] |
69
+ | 0.2002 | 13.9535 | 600 | 0.2259 | 0.9875 | 0.9876 | 0.9876 | 0.9875 | [[959, 11], [6, 389]] |
70
+ | 0.1994 | 16.2791 | 700 | 0.2255 | 0.9883 | 0.9883 | 0.9883 | 0.9883 | [[960, 10], [6, 389]] |
71
+ | 0.1997 | 18.6047 | 800 | 0.2254 | 0.9883 | 0.9883 | 0.9883 | 0.9883 | [[960, 10], [6, 389]] |
 
 
72
 
73
 
74
  ### Framework versions
checkpoint-500/config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "finetuning_task": "audio-classification",
49
+ "hidden_act": "gelu",
50
+ "hidden_dropout": 0.1,
51
+ "hidden_size": 768,
52
+ "id2label": {
53
+ "0": "crying",
54
+ "1": "no_crying"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "model_type": "hubert",
67
+ "num_attention_heads": 12,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 2,
72
+ "pad_token_id": 0,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.44.2",
75
+ "use_weighted_layer_sum": false,
76
+ "vocab_size": 32
77
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2041eca5fb53efe367a7173ea2f8134639376a8725d1d5e4b703cc5c51091e4d
3
+ size 94763496
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2841bddb32715286bfd37dc809b01547440737cf55fe0e6e2a35570191638266
3
+ size 189552570
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edf8e41f94dcf5fdfbe835965b81bc88cbb914f7ffde871d0d81b52f38954873
3
+ size 14308
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3642500630a8e812e7a47cae4b5478d99e83284cbb8a78eda0b69e2f157e5321
3
+ size 1064
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,222 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9875597008648797,
3
+ "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-500",
4
+ "epoch": 11.627906976744185,
5
+ "eval_steps": 100,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.1627906976744187,
13
+ "grad_norm": 0.9607858657836914,
14
+ "learning_rate": 1.744186046511628e-05,
15
+ "loss": 0.6124,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 2.3255813953488373,
20
+ "grad_norm": 2.0142862796783447,
21
+ "learning_rate": 2.997578868976474e-05,
22
+ "loss": 0.3124,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.3255813953488373,
27
+ "eval_accuracy": 0.9641025641025641,
28
+ "eval_confusion_matrix": [
29
+ [
30
+ 948,
31
+ 22
32
+ ],
33
+ [
34
+ 27,
35
+ 368
36
+ ]
37
+ ],
38
+ "eval_f1": 0.9640343817214121,
39
+ "eval_loss": 0.2739163041114807,
40
+ "eval_precision": 0.9639973701512163,
41
+ "eval_recall": 0.9641025641025641,
42
+ "eval_runtime": 3.48,
43
+ "eval_samples_per_second": 392.238,
44
+ "eval_steps_per_second": 6.322,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 3.488372093023256,
49
+ "grad_norm": 1.6254416704177856,
50
+ "learning_rate": 2.9496736466239833e-05,
51
+ "loss": 0.2509,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 4.651162790697675,
56
+ "grad_norm": 1.0478523969650269,
57
+ "learning_rate": 2.8422656841613377e-05,
58
+ "loss": 0.2337,
59
+ "step": 200
60
+ },
61
+ {
62
+ "epoch": 4.651162790697675,
63
+ "eval_accuracy": 0.9736263736263736,
64
+ "eval_confusion_matrix": [
65
+ [
66
+ 950,
67
+ 20
68
+ ],
69
+ [
70
+ 16,
71
+ 379
72
+ ]
73
+ ],
74
+ "eval_f1": 0.9736655604528286,
75
+ "eval_loss": 0.23845618963241577,
76
+ "eval_precision": 0.9737247242968067,
77
+ "eval_recall": 0.9736263736263736,
78
+ "eval_runtime": 3.3782,
79
+ "eval_samples_per_second": 404.066,
80
+ "eval_steps_per_second": 6.512,
81
+ "step": 200
82
+ },
83
+ {
84
+ "epoch": 5.813953488372093,
85
+ "grad_norm": 0.19202682375907898,
86
+ "learning_rate": 2.679763606999547e-05,
87
+ "loss": 0.213,
88
+ "step": 250
89
+ },
90
+ {
91
+ "epoch": 6.976744186046512,
92
+ "grad_norm": 3.5125396251678467,
93
+ "learning_rate": 2.4688374121322346e-05,
94
+ "loss": 0.2064,
95
+ "step": 300
96
+ },
97
+ {
98
+ "epoch": 6.976744186046512,
99
+ "eval_accuracy": 0.9831501831501831,
100
+ "eval_confusion_matrix": [
101
+ [
102
+ 958,
103
+ 12
104
+ ],
105
+ [
106
+ 11,
107
+ 384
108
+ ]
109
+ ],
110
+ "eval_f1": 0.9831564891330211,
111
+ "eval_loss": 0.2294875830411911,
112
+ "eval_precision": 0.983164066755398,
113
+ "eval_recall": 0.9831501831501831,
114
+ "eval_runtime": 3.421,
115
+ "eval_samples_per_second": 399.01,
116
+ "eval_steps_per_second": 6.431,
117
+ "step": 300
118
+ },
119
+ {
120
+ "epoch": 8.13953488372093,
121
+ "grad_norm": 0.06433115154504776,
122
+ "learning_rate": 2.218144694035814e-05,
123
+ "loss": 0.2044,
124
+ "step": 350
125
+ },
126
+ {
127
+ "epoch": 9.30232558139535,
128
+ "grad_norm": 0.1391066163778305,
129
+ "learning_rate": 1.9379752884171956e-05,
130
+ "loss": 0.2023,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 9.30232558139535,
135
+ "eval_accuracy": 0.9868131868131869,
136
+ "eval_confusion_matrix": [
137
+ [
138
+ 957,
139
+ 13
140
+ ],
141
+ [
142
+ 5,
143
+ 390
144
+ ]
145
+ ],
146
+ "eval_f1": 0.9868519842367506,
147
+ "eval_loss": 0.22771421074867249,
148
+ "eval_precision": 0.986971783497838,
149
+ "eval_recall": 0.9868131868131869,
150
+ "eval_runtime": 3.414,
151
+ "eval_samples_per_second": 399.822,
152
+ "eval_steps_per_second": 6.444,
153
+ "step": 400
154
+ },
155
+ {
156
+ "epoch": 10.465116279069768,
157
+ "grad_norm": 0.03317234292626381,
158
+ "learning_rate": 1.6398289196234188e-05,
159
+ "loss": 0.201,
160
+ "step": 450
161
+ },
162
+ {
163
+ "epoch": 11.627906976744185,
164
+ "grad_norm": 0.0141974575817585,
165
+ "learning_rate": 1.3359431874331888e-05,
166
+ "loss": 0.2003,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 11.627906976744185,
171
+ "eval_accuracy": 0.9875457875457876,
172
+ "eval_confusion_matrix": [
173
+ [
174
+ 960,
175
+ 10
176
+ ],
177
+ [
178
+ 7,
179
+ 388
180
+ ]
181
+ ],
182
+ "eval_f1": 0.9875597008648797,
183
+ "eval_loss": 0.22541972994804382,
184
+ "eval_precision": 0.9875850990732533,
185
+ "eval_recall": 0.9875457875457876,
186
+ "eval_runtime": 3.4267,
187
+ "eval_samples_per_second": 398.347,
188
+ "eval_steps_per_second": 6.42,
189
+ "step": 500
190
+ }
191
+ ],
192
+ "logging_steps": 50,
193
+ "max_steps": 860,
194
+ "num_input_tokens_seen": 0,
195
+ "num_train_epochs": 20,
196
+ "save_steps": 100,
197
+ "stateful_callbacks": {
198
+ "EarlyStoppingCallback": {
199
+ "args": {
200
+ "early_stopping_patience": 5,
201
+ "early_stopping_threshold": 0.001
202
+ },
203
+ "attributes": {
204
+ "early_stopping_patience_counter": 0
205
+ }
206
+ },
207
+ "TrainerControl": {
208
+ "args": {
209
+ "should_epoch_stop": false,
210
+ "should_evaluate": false,
211
+ "should_log": false,
212
+ "should_save": true,
213
+ "should_training_stop": false
214
+ },
215
+ "attributes": {}
216
+ }
217
+ },
218
+ "total_flos": 1.44446640051456e+17,
219
+ "train_batch_size": 64,
220
+ "trial_name": null,
221
+ "trial_params": null
222
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3d89420c209e2631922a2e295cec92826140b0b4b3a85040e4f26fe755df93c
3
+ size 5240
checkpoint-600/config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "finetuning_task": "audio-classification",
49
+ "hidden_act": "gelu",
50
+ "hidden_dropout": 0.1,
51
+ "hidden_size": 768,
52
+ "id2label": {
53
+ "0": "crying",
54
+ "1": "no_crying"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "model_type": "hubert",
67
+ "num_attention_heads": 12,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 2,
72
+ "pad_token_id": 0,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.44.2",
75
+ "use_weighted_layer_sum": false,
76
+ "vocab_size": 32
77
+ }
checkpoint-600/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c21555c55a0eb1aec9860b56c83a094ef848ce0579ddca6d5a2738b99a6f806
3
+ size 94763496
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7548a3259806672f97717e12300878518dfc2ef8dabff02eac23f755b765bc77
3
+ size 189552570
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:651bc52340f775bf5ca0ad060e768c602fb2b8b78ae4dc9f9e330238fb6e4131
3
+ size 14308
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbb0226b5a39dda6cf123353e8e8fb3f8a8e30903e4332b33f1dd1a021acd14c
3
+ size 1064
checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9875688610221299,
3
+ "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-600",
4
+ "epoch": 13.953488372093023,
5
+ "eval_steps": 100,
6
+ "global_step": 600,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.1627906976744187,
13
+ "grad_norm": 0.9607858657836914,
14
+ "learning_rate": 1.744186046511628e-05,
15
+ "loss": 0.6124,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 2.3255813953488373,
20
+ "grad_norm": 2.0142862796783447,
21
+ "learning_rate": 2.997578868976474e-05,
22
+ "loss": 0.3124,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.3255813953488373,
27
+ "eval_accuracy": 0.9641025641025641,
28
+ "eval_confusion_matrix": [
29
+ [
30
+ 948,
31
+ 22
32
+ ],
33
+ [
34
+ 27,
35
+ 368
36
+ ]
37
+ ],
38
+ "eval_f1": 0.9640343817214121,
39
+ "eval_loss": 0.2739163041114807,
40
+ "eval_precision": 0.9639973701512163,
41
+ "eval_recall": 0.9641025641025641,
42
+ "eval_runtime": 3.48,
43
+ "eval_samples_per_second": 392.238,
44
+ "eval_steps_per_second": 6.322,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 3.488372093023256,
49
+ "grad_norm": 1.6254416704177856,
50
+ "learning_rate": 2.9496736466239833e-05,
51
+ "loss": 0.2509,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 4.651162790697675,
56
+ "grad_norm": 1.0478523969650269,
57
+ "learning_rate": 2.8422656841613377e-05,
58
+ "loss": 0.2337,
59
+ "step": 200
60
+ },
61
+ {
62
+ "epoch": 4.651162790697675,
63
+ "eval_accuracy": 0.9736263736263736,
64
+ "eval_confusion_matrix": [
65
+ [
66
+ 950,
67
+ 20
68
+ ],
69
+ [
70
+ 16,
71
+ 379
72
+ ]
73
+ ],
74
+ "eval_f1": 0.9736655604528286,
75
+ "eval_loss": 0.23845618963241577,
76
+ "eval_precision": 0.9737247242968067,
77
+ "eval_recall": 0.9736263736263736,
78
+ "eval_runtime": 3.3782,
79
+ "eval_samples_per_second": 404.066,
80
+ "eval_steps_per_second": 6.512,
81
+ "step": 200
82
+ },
83
+ {
84
+ "epoch": 5.813953488372093,
85
+ "grad_norm": 0.19202682375907898,
86
+ "learning_rate": 2.679763606999547e-05,
87
+ "loss": 0.213,
88
+ "step": 250
89
+ },
90
+ {
91
+ "epoch": 6.976744186046512,
92
+ "grad_norm": 3.5125396251678467,
93
+ "learning_rate": 2.4688374121322346e-05,
94
+ "loss": 0.2064,
95
+ "step": 300
96
+ },
97
+ {
98
+ "epoch": 6.976744186046512,
99
+ "eval_accuracy": 0.9831501831501831,
100
+ "eval_confusion_matrix": [
101
+ [
102
+ 958,
103
+ 12
104
+ ],
105
+ [
106
+ 11,
107
+ 384
108
+ ]
109
+ ],
110
+ "eval_f1": 0.9831564891330211,
111
+ "eval_loss": 0.2294875830411911,
112
+ "eval_precision": 0.983164066755398,
113
+ "eval_recall": 0.9831501831501831,
114
+ "eval_runtime": 3.421,
115
+ "eval_samples_per_second": 399.01,
116
+ "eval_steps_per_second": 6.431,
117
+ "step": 300
118
+ },
119
+ {
120
+ "epoch": 8.13953488372093,
121
+ "grad_norm": 0.06433115154504776,
122
+ "learning_rate": 2.218144694035814e-05,
123
+ "loss": 0.2044,
124
+ "step": 350
125
+ },
126
+ {
127
+ "epoch": 9.30232558139535,
128
+ "grad_norm": 0.1391066163778305,
129
+ "learning_rate": 1.9379752884171956e-05,
130
+ "loss": 0.2023,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 9.30232558139535,
135
+ "eval_accuracy": 0.9868131868131869,
136
+ "eval_confusion_matrix": [
137
+ [
138
+ 957,
139
+ 13
140
+ ],
141
+ [
142
+ 5,
143
+ 390
144
+ ]
145
+ ],
146
+ "eval_f1": 0.9868519842367506,
147
+ "eval_loss": 0.22771421074867249,
148
+ "eval_precision": 0.986971783497838,
149
+ "eval_recall": 0.9868131868131869,
150
+ "eval_runtime": 3.414,
151
+ "eval_samples_per_second": 399.822,
152
+ "eval_steps_per_second": 6.444,
153
+ "step": 400
154
+ },
155
+ {
156
+ "epoch": 10.465116279069768,
157
+ "grad_norm": 0.03317234292626381,
158
+ "learning_rate": 1.6398289196234188e-05,
159
+ "loss": 0.201,
160
+ "step": 450
161
+ },
162
+ {
163
+ "epoch": 11.627906976744185,
164
+ "grad_norm": 0.0141974575817585,
165
+ "learning_rate": 1.3359431874331888e-05,
166
+ "loss": 0.2003,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 11.627906976744185,
171
+ "eval_accuracy": 0.9875457875457876,
172
+ "eval_confusion_matrix": [
173
+ [
174
+ 960,
175
+ 10
176
+ ],
177
+ [
178
+ 7,
179
+ 388
180
+ ]
181
+ ],
182
+ "eval_f1": 0.9875597008648797,
183
+ "eval_loss": 0.22541972994804382,
184
+ "eval_precision": 0.9875850990732533,
185
+ "eval_recall": 0.9875457875457876,
186
+ "eval_runtime": 3.4267,
187
+ "eval_samples_per_second": 398.347,
188
+ "eval_steps_per_second": 6.42,
189
+ "step": 500
190
+ },
191
+ {
192
+ "epoch": 12.790697674418604,
193
+ "grad_norm": 0.020028484985232353,
194
+ "learning_rate": 1.0387912673007252e-05,
195
+ "loss": 0.2005,
196
+ "step": 550
197
+ },
198
+ {
199
+ "epoch": 13.953488372093023,
200
+ "grad_norm": 0.013127407990396023,
201
+ "learning_rate": 7.605699412521431e-06,
202
+ "loss": 0.2002,
203
+ "step": 600
204
+ },
205
+ {
206
+ "epoch": 13.953488372093023,
207
+ "eval_accuracy": 0.9875457875457876,
208
+ "eval_confusion_matrix": [
209
+ [
210
+ 959,
211
+ 11
212
+ ],
213
+ [
214
+ 6,
215
+ 389
216
+ ]
217
+ ],
218
+ "eval_f1": 0.9875688610221299,
219
+ "eval_loss": 0.2259330302476883,
220
+ "eval_precision": 0.9876237449942113,
221
+ "eval_recall": 0.9875457875457876,
222
+ "eval_runtime": 3.3216,
223
+ "eval_samples_per_second": 410.95,
224
+ "eval_steps_per_second": 6.623,
225
+ "step": 600
226
+ }
227
+ ],
228
+ "logging_steps": 50,
229
+ "max_steps": 860,
230
+ "num_input_tokens_seen": 0,
231
+ "num_train_epochs": 20,
232
+ "save_steps": 100,
233
+ "stateful_callbacks": {
234
+ "EarlyStoppingCallback": {
235
+ "args": {
236
+ "early_stopping_patience": 5,
237
+ "early_stopping_threshold": 0.001
238
+ },
239
+ "attributes": {
240
+ "early_stopping_patience_counter": 0
241
+ }
242
+ },
243
+ "TrainerControl": {
244
+ "args": {
245
+ "should_epoch_stop": false,
246
+ "should_evaluate": false,
247
+ "should_log": false,
248
+ "should_save": true,
249
+ "should_training_stop": false
250
+ },
251
+ "attributes": {}
252
+ }
253
+ },
254
+ "total_flos": 1.73355980821248e+17,
255
+ "train_batch_size": 64,
256
+ "trial_name": null,
257
+ "trial_params": null
258
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3d89420c209e2631922a2e295cec92826140b0b4b3a85040e4f26fe755df93c
3
+ size 5240
checkpoint-700/config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "finetuning_task": "audio-classification",
49
+ "hidden_act": "gelu",
50
+ "hidden_dropout": 0.1,
51
+ "hidden_size": 768,
52
+ "id2label": {
53
+ "0": "crying",
54
+ "1": "no_crying"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "model_type": "hubert",
67
+ "num_attention_heads": 12,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 2,
72
+ "pad_token_id": 0,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.44.2",
75
+ "use_weighted_layer_sum": false,
76
+ "vocab_size": 32
77
+ }
checkpoint-700/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc9d6a3efaa15139c296f80cc6ffb6fb613e832c8d723e4e3b6b9262917a4d3c
3
+ size 94763496
checkpoint-700/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98e2c90128a1e2ea359948ddde461ddbd528fd38eb7b3d98c943f471de53b0fd
3
+ size 189552570
checkpoint-700/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b58dca5fb1b7169134a579fbefe18f63835f88838b2ef059378fcdd3c9cecd5f
3
+ size 14308
checkpoint-700/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9763cc72a9be84b93ee3f8307b04709caa68b45aea1c765c4087e9b0f2cec06b
3
+ size 1064
checkpoint-700/trainer_state.json ADDED
@@ -0,0 +1,294 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9882958046457017,
3
+ "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-700",
4
+ "epoch": 16.27906976744186,
5
+ "eval_steps": 100,
6
+ "global_step": 700,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.1627906976744187,
13
+ "grad_norm": 0.9607858657836914,
14
+ "learning_rate": 1.744186046511628e-05,
15
+ "loss": 0.6124,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 2.3255813953488373,
20
+ "grad_norm": 2.0142862796783447,
21
+ "learning_rate": 2.997578868976474e-05,
22
+ "loss": 0.3124,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.3255813953488373,
27
+ "eval_accuracy": 0.9641025641025641,
28
+ "eval_confusion_matrix": [
29
+ [
30
+ 948,
31
+ 22
32
+ ],
33
+ [
34
+ 27,
35
+ 368
36
+ ]
37
+ ],
38
+ "eval_f1": 0.9640343817214121,
39
+ "eval_loss": 0.2739163041114807,
40
+ "eval_precision": 0.9639973701512163,
41
+ "eval_recall": 0.9641025641025641,
42
+ "eval_runtime": 3.48,
43
+ "eval_samples_per_second": 392.238,
44
+ "eval_steps_per_second": 6.322,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 3.488372093023256,
49
+ "grad_norm": 1.6254416704177856,
50
+ "learning_rate": 2.9496736466239833e-05,
51
+ "loss": 0.2509,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 4.651162790697675,
56
+ "grad_norm": 1.0478523969650269,
57
+ "learning_rate": 2.8422656841613377e-05,
58
+ "loss": 0.2337,
59
+ "step": 200
60
+ },
61
+ {
62
+ "epoch": 4.651162790697675,
63
+ "eval_accuracy": 0.9736263736263736,
64
+ "eval_confusion_matrix": [
65
+ [
66
+ 950,
67
+ 20
68
+ ],
69
+ [
70
+ 16,
71
+ 379
72
+ ]
73
+ ],
74
+ "eval_f1": 0.9736655604528286,
75
+ "eval_loss": 0.23845618963241577,
76
+ "eval_precision": 0.9737247242968067,
77
+ "eval_recall": 0.9736263736263736,
78
+ "eval_runtime": 3.3782,
79
+ "eval_samples_per_second": 404.066,
80
+ "eval_steps_per_second": 6.512,
81
+ "step": 200
82
+ },
83
+ {
84
+ "epoch": 5.813953488372093,
85
+ "grad_norm": 0.19202682375907898,
86
+ "learning_rate": 2.679763606999547e-05,
87
+ "loss": 0.213,
88
+ "step": 250
89
+ },
90
+ {
91
+ "epoch": 6.976744186046512,
92
+ "grad_norm": 3.5125396251678467,
93
+ "learning_rate": 2.4688374121322346e-05,
94
+ "loss": 0.2064,
95
+ "step": 300
96
+ },
97
+ {
98
+ "epoch": 6.976744186046512,
99
+ "eval_accuracy": 0.9831501831501831,
100
+ "eval_confusion_matrix": [
101
+ [
102
+ 958,
103
+ 12
104
+ ],
105
+ [
106
+ 11,
107
+ 384
108
+ ]
109
+ ],
110
+ "eval_f1": 0.9831564891330211,
111
+ "eval_loss": 0.2294875830411911,
112
+ "eval_precision": 0.983164066755398,
113
+ "eval_recall": 0.9831501831501831,
114
+ "eval_runtime": 3.421,
115
+ "eval_samples_per_second": 399.01,
116
+ "eval_steps_per_second": 6.431,
117
+ "step": 300
118
+ },
119
+ {
120
+ "epoch": 8.13953488372093,
121
+ "grad_norm": 0.06433115154504776,
122
+ "learning_rate": 2.218144694035814e-05,
123
+ "loss": 0.2044,
124
+ "step": 350
125
+ },
126
+ {
127
+ "epoch": 9.30232558139535,
128
+ "grad_norm": 0.1391066163778305,
129
+ "learning_rate": 1.9379752884171956e-05,
130
+ "loss": 0.2023,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 9.30232558139535,
135
+ "eval_accuracy": 0.9868131868131869,
136
+ "eval_confusion_matrix": [
137
+ [
138
+ 957,
139
+ 13
140
+ ],
141
+ [
142
+ 5,
143
+ 390
144
+ ]
145
+ ],
146
+ "eval_f1": 0.9868519842367506,
147
+ "eval_loss": 0.22771421074867249,
148
+ "eval_precision": 0.986971783497838,
149
+ "eval_recall": 0.9868131868131869,
150
+ "eval_runtime": 3.414,
151
+ "eval_samples_per_second": 399.822,
152
+ "eval_steps_per_second": 6.444,
153
+ "step": 400
154
+ },
155
+ {
156
+ "epoch": 10.465116279069768,
157
+ "grad_norm": 0.03317234292626381,
158
+ "learning_rate": 1.6398289196234188e-05,
159
+ "loss": 0.201,
160
+ "step": 450
161
+ },
162
+ {
163
+ "epoch": 11.627906976744185,
164
+ "grad_norm": 0.0141974575817585,
165
+ "learning_rate": 1.3359431874331888e-05,
166
+ "loss": 0.2003,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 11.627906976744185,
171
+ "eval_accuracy": 0.9875457875457876,
172
+ "eval_confusion_matrix": [
173
+ [
174
+ 960,
175
+ 10
176
+ ],
177
+ [
178
+ 7,
179
+ 388
180
+ ]
181
+ ],
182
+ "eval_f1": 0.9875597008648797,
183
+ "eval_loss": 0.22541972994804382,
184
+ "eval_precision": 0.9875850990732533,
185
+ "eval_recall": 0.9875457875457876,
186
+ "eval_runtime": 3.4267,
187
+ "eval_samples_per_second": 398.347,
188
+ "eval_steps_per_second": 6.42,
189
+ "step": 500
190
+ },
191
+ {
192
+ "epoch": 12.790697674418604,
193
+ "grad_norm": 0.020028484985232353,
194
+ "learning_rate": 1.0387912673007252e-05,
195
+ "loss": 0.2005,
196
+ "step": 550
197
+ },
198
+ {
199
+ "epoch": 13.953488372093023,
200
+ "grad_norm": 0.013127407990396023,
201
+ "learning_rate": 7.605699412521431e-06,
202
+ "loss": 0.2002,
203
+ "step": 600
204
+ },
205
+ {
206
+ "epoch": 13.953488372093023,
207
+ "eval_accuracy": 0.9875457875457876,
208
+ "eval_confusion_matrix": [
209
+ [
210
+ 959,
211
+ 11
212
+ ],
213
+ [
214
+ 6,
215
+ 389
216
+ ]
217
+ ],
218
+ "eval_f1": 0.9875688610221299,
219
+ "eval_loss": 0.2259330302476883,
220
+ "eval_precision": 0.9876237449942113,
221
+ "eval_recall": 0.9875457875457876,
222
+ "eval_runtime": 3.3216,
223
+ "eval_samples_per_second": 410.95,
224
+ "eval_steps_per_second": 6.623,
225
+ "step": 600
226
+ },
227
+ {
228
+ "epoch": 15.116279069767442,
229
+ "grad_norm": 0.016032038256525993,
230
+ "learning_rate": 5.126989735188782e-06,
231
+ "loss": 0.2012,
232
+ "step": 650
233
+ },
234
+ {
235
+ "epoch": 16.27906976744186,
236
+ "grad_norm": 0.012081542983651161,
237
+ "learning_rate": 3.0535237934100274e-06,
238
+ "loss": 0.1994,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 16.27906976744186,
243
+ "eval_accuracy": 0.9882783882783883,
244
+ "eval_confusion_matrix": [
245
+ [
246
+ 960,
247
+ 10
248
+ ],
249
+ [
250
+ 6,
251
+ 389
252
+ ]
253
+ ],
254
+ "eval_f1": 0.9882958046457017,
255
+ "eval_loss": 0.22554340958595276,
256
+ "eval_precision": 0.9883336307004172,
257
+ "eval_recall": 0.9882783882783883,
258
+ "eval_runtime": 3.3839,
259
+ "eval_samples_per_second": 403.383,
260
+ "eval_steps_per_second": 6.501,
261
+ "step": 700
262
+ }
263
+ ],
264
+ "logging_steps": 50,
265
+ "max_steps": 860,
266
+ "num_input_tokens_seen": 0,
267
+ "num_train_epochs": 20,
268
+ "save_steps": 100,
269
+ "stateful_callbacks": {
270
+ "EarlyStoppingCallback": {
271
+ "args": {
272
+ "early_stopping_patience": 5,
273
+ "early_stopping_threshold": 0.001
274
+ },
275
+ "attributes": {
276
+ "early_stopping_patience_counter": 0
277
+ }
278
+ },
279
+ "TrainerControl": {
280
+ "args": {
281
+ "should_epoch_stop": false,
282
+ "should_evaluate": false,
283
+ "should_log": false,
284
+ "should_save": true,
285
+ "should_training_stop": false
286
+ },
287
+ "attributes": {}
288
+ }
289
+ },
290
+ "total_flos": 2.02165257793536e+17,
291
+ "train_batch_size": 64,
292
+ "trial_name": null,
293
+ "trial_params": null
294
+ }
checkpoint-700/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3d89420c209e2631922a2e295cec92826140b0b4b3a85040e4f26fe755df93c
3
+ size 5240
checkpoint-800/config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "finetuning_task": "audio-classification",
49
+ "hidden_act": "gelu",
50
+ "hidden_dropout": 0.1,
51
+ "hidden_size": 768,
52
+ "id2label": {
53
+ "0": "crying",
54
+ "1": "no_crying"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "model_type": "hubert",
67
+ "num_attention_heads": 12,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 2,
72
+ "pad_token_id": 0,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.44.2",
75
+ "use_weighted_layer_sum": false,
76
+ "vocab_size": 32
77
+ }
checkpoint-800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ca2c8b3045e83eee0dd3997a0d466b85f518369f3117e39e4afe419a6cab06b
3
+ size 94763496
checkpoint-800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48fe4b3c8426713ecb377f192fbe2278ecf2710151ea3b3a8d8e5710d4c0dd32
3
+ size 189552570
checkpoint-800/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b21d10b3ec1e745fe106985b1f76b00d319efb8e992b1ee18f7aa51bf01b984
3
+ size 14308
checkpoint-800/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e085557df30c0fd508f88a56be4593075bb2341a5f76662bf774b850aa237abb
3
+ size 1064
checkpoint-800/trainer_state.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9882958046457017,
3
+ "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-700",
4
+ "epoch": 18.6046511627907,
5
+ "eval_steps": 100,
6
+ "global_step": 800,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.1627906976744187,
13
+ "grad_norm": 0.9607858657836914,
14
+ "learning_rate": 1.744186046511628e-05,
15
+ "loss": 0.6124,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 2.3255813953488373,
20
+ "grad_norm": 2.0142862796783447,
21
+ "learning_rate": 2.997578868976474e-05,
22
+ "loss": 0.3124,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.3255813953488373,
27
+ "eval_accuracy": 0.9641025641025641,
28
+ "eval_confusion_matrix": [
29
+ [
30
+ 948,
31
+ 22
32
+ ],
33
+ [
34
+ 27,
35
+ 368
36
+ ]
37
+ ],
38
+ "eval_f1": 0.9640343817214121,
39
+ "eval_loss": 0.2739163041114807,
40
+ "eval_precision": 0.9639973701512163,
41
+ "eval_recall": 0.9641025641025641,
42
+ "eval_runtime": 3.48,
43
+ "eval_samples_per_second": 392.238,
44
+ "eval_steps_per_second": 6.322,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 3.488372093023256,
49
+ "grad_norm": 1.6254416704177856,
50
+ "learning_rate": 2.9496736466239833e-05,
51
+ "loss": 0.2509,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 4.651162790697675,
56
+ "grad_norm": 1.0478523969650269,
57
+ "learning_rate": 2.8422656841613377e-05,
58
+ "loss": 0.2337,
59
+ "step": 200
60
+ },
61
+ {
62
+ "epoch": 4.651162790697675,
63
+ "eval_accuracy": 0.9736263736263736,
64
+ "eval_confusion_matrix": [
65
+ [
66
+ 950,
67
+ 20
68
+ ],
69
+ [
70
+ 16,
71
+ 379
72
+ ]
73
+ ],
74
+ "eval_f1": 0.9736655604528286,
75
+ "eval_loss": 0.23845618963241577,
76
+ "eval_precision": 0.9737247242968067,
77
+ "eval_recall": 0.9736263736263736,
78
+ "eval_runtime": 3.3782,
79
+ "eval_samples_per_second": 404.066,
80
+ "eval_steps_per_second": 6.512,
81
+ "step": 200
82
+ },
83
+ {
84
+ "epoch": 5.813953488372093,
85
+ "grad_norm": 0.19202682375907898,
86
+ "learning_rate": 2.679763606999547e-05,
87
+ "loss": 0.213,
88
+ "step": 250
89
+ },
90
+ {
91
+ "epoch": 6.976744186046512,
92
+ "grad_norm": 3.5125396251678467,
93
+ "learning_rate": 2.4688374121322346e-05,
94
+ "loss": 0.2064,
95
+ "step": 300
96
+ },
97
+ {
98
+ "epoch": 6.976744186046512,
99
+ "eval_accuracy": 0.9831501831501831,
100
+ "eval_confusion_matrix": [
101
+ [
102
+ 958,
103
+ 12
104
+ ],
105
+ [
106
+ 11,
107
+ 384
108
+ ]
109
+ ],
110
+ "eval_f1": 0.9831564891330211,
111
+ "eval_loss": 0.2294875830411911,
112
+ "eval_precision": 0.983164066755398,
113
+ "eval_recall": 0.9831501831501831,
114
+ "eval_runtime": 3.421,
115
+ "eval_samples_per_second": 399.01,
116
+ "eval_steps_per_second": 6.431,
117
+ "step": 300
118
+ },
119
+ {
120
+ "epoch": 8.13953488372093,
121
+ "grad_norm": 0.06433115154504776,
122
+ "learning_rate": 2.218144694035814e-05,
123
+ "loss": 0.2044,
124
+ "step": 350
125
+ },
126
+ {
127
+ "epoch": 9.30232558139535,
128
+ "grad_norm": 0.1391066163778305,
129
+ "learning_rate": 1.9379752884171956e-05,
130
+ "loss": 0.2023,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 9.30232558139535,
135
+ "eval_accuracy": 0.9868131868131869,
136
+ "eval_confusion_matrix": [
137
+ [
138
+ 957,
139
+ 13
140
+ ],
141
+ [
142
+ 5,
143
+ 390
144
+ ]
145
+ ],
146
+ "eval_f1": 0.9868519842367506,
147
+ "eval_loss": 0.22771421074867249,
148
+ "eval_precision": 0.986971783497838,
149
+ "eval_recall": 0.9868131868131869,
150
+ "eval_runtime": 3.414,
151
+ "eval_samples_per_second": 399.822,
152
+ "eval_steps_per_second": 6.444,
153
+ "step": 400
154
+ },
155
+ {
156
+ "epoch": 10.465116279069768,
157
+ "grad_norm": 0.03317234292626381,
158
+ "learning_rate": 1.6398289196234188e-05,
159
+ "loss": 0.201,
160
+ "step": 450
161
+ },
162
+ {
163
+ "epoch": 11.627906976744185,
164
+ "grad_norm": 0.0141974575817585,
165
+ "learning_rate": 1.3359431874331888e-05,
166
+ "loss": 0.2003,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 11.627906976744185,
171
+ "eval_accuracy": 0.9875457875457876,
172
+ "eval_confusion_matrix": [
173
+ [
174
+ 960,
175
+ 10
176
+ ],
177
+ [
178
+ 7,
179
+ 388
180
+ ]
181
+ ],
182
+ "eval_f1": 0.9875597008648797,
183
+ "eval_loss": 0.22541972994804382,
184
+ "eval_precision": 0.9875850990732533,
185
+ "eval_recall": 0.9875457875457876,
186
+ "eval_runtime": 3.4267,
187
+ "eval_samples_per_second": 398.347,
188
+ "eval_steps_per_second": 6.42,
189
+ "step": 500
190
+ },
191
+ {
192
+ "epoch": 12.790697674418604,
193
+ "grad_norm": 0.020028484985232353,
194
+ "learning_rate": 1.0387912673007252e-05,
195
+ "loss": 0.2005,
196
+ "step": 550
197
+ },
198
+ {
199
+ "epoch": 13.953488372093023,
200
+ "grad_norm": 0.013127407990396023,
201
+ "learning_rate": 7.605699412521431e-06,
202
+ "loss": 0.2002,
203
+ "step": 600
204
+ },
205
+ {
206
+ "epoch": 13.953488372093023,
207
+ "eval_accuracy": 0.9875457875457876,
208
+ "eval_confusion_matrix": [
209
+ [
210
+ 959,
211
+ 11
212
+ ],
213
+ [
214
+ 6,
215
+ 389
216
+ ]
217
+ ],
218
+ "eval_f1": 0.9875688610221299,
219
+ "eval_loss": 0.2259330302476883,
220
+ "eval_precision": 0.9876237449942113,
221
+ "eval_recall": 0.9875457875457876,
222
+ "eval_runtime": 3.3216,
223
+ "eval_samples_per_second": 410.95,
224
+ "eval_steps_per_second": 6.623,
225
+ "step": 600
226
+ },
227
+ {
228
+ "epoch": 15.116279069767442,
229
+ "grad_norm": 0.016032038256525993,
230
+ "learning_rate": 5.126989735188782e-06,
231
+ "loss": 0.2012,
232
+ "step": 650
233
+ },
234
+ {
235
+ "epoch": 16.27906976744186,
236
+ "grad_norm": 0.012081542983651161,
237
+ "learning_rate": 3.0535237934100274e-06,
238
+ "loss": 0.1994,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 16.27906976744186,
243
+ "eval_accuracy": 0.9882783882783883,
244
+ "eval_confusion_matrix": [
245
+ [
246
+ 960,
247
+ 10
248
+ ],
249
+ [
250
+ 6,
251
+ 389
252
+ ]
253
+ ],
254
+ "eval_f1": 0.9882958046457017,
255
+ "eval_loss": 0.22554340958595276,
256
+ "eval_precision": 0.9883336307004172,
257
+ "eval_recall": 0.9882783882783883,
258
+ "eval_runtime": 3.3839,
259
+ "eval_samples_per_second": 403.383,
260
+ "eval_steps_per_second": 6.501,
261
+ "step": 700
262
+ },
263
+ {
264
+ "epoch": 17.441860465116278,
265
+ "grad_norm": 0.015530905686318874,
266
+ "learning_rate": 1.470408262986479e-06,
267
+ "loss": 0.2008,
268
+ "step": 750
269
+ },
270
+ {
271
+ "epoch": 18.6046511627907,
272
+ "grad_norm": 0.03649911284446716,
273
+ "learning_rate": 4.4262308763892087e-07,
274
+ "loss": 0.1997,
275
+ "step": 800
276
+ },
277
+ {
278
+ "epoch": 18.6046511627907,
279
+ "eval_accuracy": 0.9882783882783883,
280
+ "eval_confusion_matrix": [
281
+ [
282
+ 960,
283
+ 10
284
+ ],
285
+ [
286
+ 6,
287
+ 389
288
+ ]
289
+ ],
290
+ "eval_f1": 0.9882958046457017,
291
+ "eval_loss": 0.22540099918842316,
292
+ "eval_precision": 0.9883336307004172,
293
+ "eval_recall": 0.9882783882783883,
294
+ "eval_runtime": 3.38,
295
+ "eval_samples_per_second": 403.852,
296
+ "eval_steps_per_second": 6.509,
297
+ "step": 800
298
+ }
299
+ ],
300
+ "logging_steps": 50,
301
+ "max_steps": 860,
302
+ "num_input_tokens_seen": 0,
303
+ "num_train_epochs": 20,
304
+ "save_steps": 100,
305
+ "stateful_callbacks": {
306
+ "EarlyStoppingCallback": {
307
+ "args": {
308
+ "early_stopping_patience": 5,
309
+ "early_stopping_threshold": 0.001
310
+ },
311
+ "attributes": {
312
+ "early_stopping_patience_counter": 0
313
+ }
314
+ },
315
+ "TrainerControl": {
316
+ "args": {
317
+ "should_epoch_stop": false,
318
+ "should_evaluate": false,
319
+ "should_log": false,
320
+ "should_save": true,
321
+ "should_training_stop": false
322
+ },
323
+ "attributes": {}
324
+ }
325
+ },
326
+ "total_flos": 2.31074598563328e+17,
327
+ "train_batch_size": 64,
328
+ "trial_name": null,
329
+ "trial_params": null
330
+ }
checkpoint-800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3d89420c209e2631922a2e295cec92826140b0b4b3a85040e4f26fe755df93c
3
+ size 5240
checkpoint-860/config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "finetuning_task": "audio-classification",
49
+ "hidden_act": "gelu",
50
+ "hidden_dropout": 0.1,
51
+ "hidden_size": 768,
52
+ "id2label": {
53
+ "0": "crying",
54
+ "1": "no_crying"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "model_type": "hubert",
67
+ "num_attention_heads": 12,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 2,
72
+ "pad_token_id": 0,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.44.2",
75
+ "use_weighted_layer_sum": false,
76
+ "vocab_size": 32
77
+ }
checkpoint-860/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39d7ea4d47fbee3393f5e749620e0b4a578cea2449907bfe0372d868b5f4b30a
3
+ size 94763496
checkpoint-860/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bfa4331625c52dd28223fd3b2cba322b7836ef0875a75d29701a3fd08df747a
3
+ size 189552570
checkpoint-860/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea8e1549744101389fee1cb4923c7729bf266144965aa751df09dad62532a0fd
3
+ size 14308
checkpoint-860/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3305b0d94ed5c5807ba878e31359e591ea9ddd9f01e4c81eb836abb2e249bfab
3
+ size 1064
checkpoint-860/trainer_state.json ADDED
@@ -0,0 +1,337 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9882958046457017,
3
+ "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-700",
4
+ "epoch": 20.0,
5
+ "eval_steps": 100,
6
+ "global_step": 860,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.1627906976744187,
13
+ "grad_norm": 0.9607858657836914,
14
+ "learning_rate": 1.744186046511628e-05,
15
+ "loss": 0.6124,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 2.3255813953488373,
20
+ "grad_norm": 2.0142862796783447,
21
+ "learning_rate": 2.997578868976474e-05,
22
+ "loss": 0.3124,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.3255813953488373,
27
+ "eval_accuracy": 0.9641025641025641,
28
+ "eval_confusion_matrix": [
29
+ [
30
+ 948,
31
+ 22
32
+ ],
33
+ [
34
+ 27,
35
+ 368
36
+ ]
37
+ ],
38
+ "eval_f1": 0.9640343817214121,
39
+ "eval_loss": 0.2739163041114807,
40
+ "eval_precision": 0.9639973701512163,
41
+ "eval_recall": 0.9641025641025641,
42
+ "eval_runtime": 3.48,
43
+ "eval_samples_per_second": 392.238,
44
+ "eval_steps_per_second": 6.322,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 3.488372093023256,
49
+ "grad_norm": 1.6254416704177856,
50
+ "learning_rate": 2.9496736466239833e-05,
51
+ "loss": 0.2509,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 4.651162790697675,
56
+ "grad_norm": 1.0478523969650269,
57
+ "learning_rate": 2.8422656841613377e-05,
58
+ "loss": 0.2337,
59
+ "step": 200
60
+ },
61
+ {
62
+ "epoch": 4.651162790697675,
63
+ "eval_accuracy": 0.9736263736263736,
64
+ "eval_confusion_matrix": [
65
+ [
66
+ 950,
67
+ 20
68
+ ],
69
+ [
70
+ 16,
71
+ 379
72
+ ]
73
+ ],
74
+ "eval_f1": 0.9736655604528286,
75
+ "eval_loss": 0.23845618963241577,
76
+ "eval_precision": 0.9737247242968067,
77
+ "eval_recall": 0.9736263736263736,
78
+ "eval_runtime": 3.3782,
79
+ "eval_samples_per_second": 404.066,
80
+ "eval_steps_per_second": 6.512,
81
+ "step": 200
82
+ },
83
+ {
84
+ "epoch": 5.813953488372093,
85
+ "grad_norm": 0.19202682375907898,
86
+ "learning_rate": 2.679763606999547e-05,
87
+ "loss": 0.213,
88
+ "step": 250
89
+ },
90
+ {
91
+ "epoch": 6.976744186046512,
92
+ "grad_norm": 3.5125396251678467,
93
+ "learning_rate": 2.4688374121322346e-05,
94
+ "loss": 0.2064,
95
+ "step": 300
96
+ },
97
+ {
98
+ "epoch": 6.976744186046512,
99
+ "eval_accuracy": 0.9831501831501831,
100
+ "eval_confusion_matrix": [
101
+ [
102
+ 958,
103
+ 12
104
+ ],
105
+ [
106
+ 11,
107
+ 384
108
+ ]
109
+ ],
110
+ "eval_f1": 0.9831564891330211,
111
+ "eval_loss": 0.2294875830411911,
112
+ "eval_precision": 0.983164066755398,
113
+ "eval_recall": 0.9831501831501831,
114
+ "eval_runtime": 3.421,
115
+ "eval_samples_per_second": 399.01,
116
+ "eval_steps_per_second": 6.431,
117
+ "step": 300
118
+ },
119
+ {
120
+ "epoch": 8.13953488372093,
121
+ "grad_norm": 0.06433115154504776,
122
+ "learning_rate": 2.218144694035814e-05,
123
+ "loss": 0.2044,
124
+ "step": 350
125
+ },
126
+ {
127
+ "epoch": 9.30232558139535,
128
+ "grad_norm": 0.1391066163778305,
129
+ "learning_rate": 1.9379752884171956e-05,
130
+ "loss": 0.2023,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 9.30232558139535,
135
+ "eval_accuracy": 0.9868131868131869,
136
+ "eval_confusion_matrix": [
137
+ [
138
+ 957,
139
+ 13
140
+ ],
141
+ [
142
+ 5,
143
+ 390
144
+ ]
145
+ ],
146
+ "eval_f1": 0.9868519842367506,
147
+ "eval_loss": 0.22771421074867249,
148
+ "eval_precision": 0.986971783497838,
149
+ "eval_recall": 0.9868131868131869,
150
+ "eval_runtime": 3.414,
151
+ "eval_samples_per_second": 399.822,
152
+ "eval_steps_per_second": 6.444,
153
+ "step": 400
154
+ },
155
+ {
156
+ "epoch": 10.465116279069768,
157
+ "grad_norm": 0.03317234292626381,
158
+ "learning_rate": 1.6398289196234188e-05,
159
+ "loss": 0.201,
160
+ "step": 450
161
+ },
162
+ {
163
+ "epoch": 11.627906976744185,
164
+ "grad_norm": 0.0141974575817585,
165
+ "learning_rate": 1.3359431874331888e-05,
166
+ "loss": 0.2003,
167
+ "step": 500
168
+ },
169
+ {
170
+ "epoch": 11.627906976744185,
171
+ "eval_accuracy": 0.9875457875457876,
172
+ "eval_confusion_matrix": [
173
+ [
174
+ 960,
175
+ 10
176
+ ],
177
+ [
178
+ 7,
179
+ 388
180
+ ]
181
+ ],
182
+ "eval_f1": 0.9875597008648797,
183
+ "eval_loss": 0.22541972994804382,
184
+ "eval_precision": 0.9875850990732533,
185
+ "eval_recall": 0.9875457875457876,
186
+ "eval_runtime": 3.4267,
187
+ "eval_samples_per_second": 398.347,
188
+ "eval_steps_per_second": 6.42,
189
+ "step": 500
190
+ },
191
+ {
192
+ "epoch": 12.790697674418604,
193
+ "grad_norm": 0.020028484985232353,
194
+ "learning_rate": 1.0387912673007252e-05,
195
+ "loss": 0.2005,
196
+ "step": 550
197
+ },
198
+ {
199
+ "epoch": 13.953488372093023,
200
+ "grad_norm": 0.013127407990396023,
201
+ "learning_rate": 7.605699412521431e-06,
202
+ "loss": 0.2002,
203
+ "step": 600
204
+ },
205
+ {
206
+ "epoch": 13.953488372093023,
207
+ "eval_accuracy": 0.9875457875457876,
208
+ "eval_confusion_matrix": [
209
+ [
210
+ 959,
211
+ 11
212
+ ],
213
+ [
214
+ 6,
215
+ 389
216
+ ]
217
+ ],
218
+ "eval_f1": 0.9875688610221299,
219
+ "eval_loss": 0.2259330302476883,
220
+ "eval_precision": 0.9876237449942113,
221
+ "eval_recall": 0.9875457875457876,
222
+ "eval_runtime": 3.3216,
223
+ "eval_samples_per_second": 410.95,
224
+ "eval_steps_per_second": 6.623,
225
+ "step": 600
226
+ },
227
+ {
228
+ "epoch": 15.116279069767442,
229
+ "grad_norm": 0.016032038256525993,
230
+ "learning_rate": 5.126989735188782e-06,
231
+ "loss": 0.2012,
232
+ "step": 650
233
+ },
234
+ {
235
+ "epoch": 16.27906976744186,
236
+ "grad_norm": 0.012081542983651161,
237
+ "learning_rate": 3.0535237934100274e-06,
238
+ "loss": 0.1994,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 16.27906976744186,
243
+ "eval_accuracy": 0.9882783882783883,
244
+ "eval_confusion_matrix": [
245
+ [
246
+ 960,
247
+ 10
248
+ ],
249
+ [
250
+ 6,
251
+ 389
252
+ ]
253
+ ],
254
+ "eval_f1": 0.9882958046457017,
255
+ "eval_loss": 0.22554340958595276,
256
+ "eval_precision": 0.9883336307004172,
257
+ "eval_recall": 0.9882783882783883,
258
+ "eval_runtime": 3.3839,
259
+ "eval_samples_per_second": 403.383,
260
+ "eval_steps_per_second": 6.501,
261
+ "step": 700
262
+ },
263
+ {
264
+ "epoch": 17.441860465116278,
265
+ "grad_norm": 0.015530905686318874,
266
+ "learning_rate": 1.470408262986479e-06,
267
+ "loss": 0.2008,
268
+ "step": 750
269
+ },
270
+ {
271
+ "epoch": 18.6046511627907,
272
+ "grad_norm": 0.03649911284446716,
273
+ "learning_rate": 4.4262308763892087e-07,
274
+ "loss": 0.1997,
275
+ "step": 800
276
+ },
277
+ {
278
+ "epoch": 18.6046511627907,
279
+ "eval_accuracy": 0.9882783882783883,
280
+ "eval_confusion_matrix": [
281
+ [
282
+ 960,
283
+ 10
284
+ ],
285
+ [
286
+ 6,
287
+ 389
288
+ ]
289
+ ],
290
+ "eval_f1": 0.9882958046457017,
291
+ "eval_loss": 0.22540099918842316,
292
+ "eval_precision": 0.9883336307004172,
293
+ "eval_recall": 0.9882783882783883,
294
+ "eval_runtime": 3.38,
295
+ "eval_samples_per_second": 403.852,
296
+ "eval_steps_per_second": 6.509,
297
+ "step": 800
298
+ },
299
+ {
300
+ "epoch": 19.767441860465116,
301
+ "grad_norm": 0.17561723291873932,
302
+ "learning_rate": 1.2354337535766536e-08,
303
+ "loss": 0.2008,
304
+ "step": 850
305
+ }
306
+ ],
307
+ "logging_steps": 50,
308
+ "max_steps": 860,
309
+ "num_input_tokens_seen": 0,
310
+ "num_train_epochs": 20,
311
+ "save_steps": 100,
312
+ "stateful_callbacks": {
313
+ "EarlyStoppingCallback": {
314
+ "args": {
315
+ "early_stopping_patience": 5,
316
+ "early_stopping_threshold": 0.001
317
+ },
318
+ "attributes": {
319
+ "early_stopping_patience_counter": 0
320
+ }
321
+ },
322
+ "TrainerControl": {
323
+ "args": {
324
+ "should_epoch_stop": false,
325
+ "should_evaluate": false,
326
+ "should_log": false,
327
+ "should_save": true,
328
+ "should_training_stop": true
329
+ },
330
+ "attributes": {}
331
+ }
332
+ },
333
+ "total_flos": 2.483401519872e+17,
334
+ "train_batch_size": 64,
335
+ "trial_name": null,
336
+ "trial_params": null
337
+ }
checkpoint-860/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3d89420c209e2631922a2e295cec92826140b0b4b3a85040e4f26fe755df93c
3
+ size 5240
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3fefc9fc9e0c6f521d52f6944550be09ad08cf2a406465b2bad96d43e8e656a9
3
  size 94763496
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc9d6a3efaa15139c296f80cc6ffb6fb613e832c8d723e4e3b6b9262917a4d3c
3
  size 94763496
runs/Sep11_15-20-39_ubumarcos/events.out.tfevents.1726060840.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3457d6d32eee966361cd1a5e55bfc41aea0b8448b79e8dc0419ad9f56364d56
3
+ size 5795
runs/Sep11_15-21-14_ubumarcos/events.out.tfevents.1726060876.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2677a6a2bf7d8a64d0c79c173ee4f873dd3a521604a432950bae62def6d8e06e
3
+ size 7184
runs/Sep11_15-25-59_ubumarcos/events.out.tfevents.1726061160.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8dbe9a7696e2700e9ae245d9bd2b3e8f9ce8e4c4724beed7069fcf4b8827c826
3
+ size 10773
runs/Sep11_15-25-59_ubumarcos/events.out.tfevents.1726061977.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea600bf545d3d4626fd88b275f082cd15a4449c22b38bdc432c40f465a710247
3
+ size 503
runs/Sep11_15-42-12_ubumarcos/events.out.tfevents.1726062133.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4204743a7822eb72e21292b34c8728a7257c7e6624882cef2790dc8e48c7dee0
3
+ size 7647
runs/Sep11_15-50-04_ubumarcos/events.out.tfevents.1726062605.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae11e51efaad3daa9c0afdc2f11dbd3f7305426132549ec41a4a4a274c18c988
3
+ size 5806
runs/Sep11_15-50-31_ubumarcos/events.out.tfevents.1726062632.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc9e10b14d9386a106cfe9f147897a38340fa2232ba11190681651623aa3eb1a
3
+ size 5804
runs/Sep11_15-50-57_ubumarcos/events.out.tfevents.1726062658.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e350751289634829a9fd460749053434acd1af6a93ed53f32d40ba013fa2170
3
+ size 5806
runs/Sep11_15-51-52_ubumarcos/events.out.tfevents.1726062713.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d152b790394ef1654f41e83eee46f69ad24e2c6949b44a7d10b165f8445e42a7
3
+ size 5807
runs/Sep11_15-52-20_ubumarcos/events.out.tfevents.1726062741.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54b53922a640a17f8f76424c75f29092341a456c1ab191c88af0a378c2b04c97
3
+ size 5806
runs/Sep11_15-52-43_ubumarcos/events.out.tfevents.1726062764.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:948a1e01a387f8ee74909d6ff171610a41577bf81caef495d41daecdf546b406
3
+ size 5807
runs/Sep11_15-54-06_ubumarcos/events.out.tfevents.1726062847.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:012e4be1fb2dcb0201aad7f28222fcb79340b9cf0fc188a95c1eb9d0e43a8b33
3
+ size 5807
runs/Sep11_15-54-19_ubumarcos/events.out.tfevents.1726062860.ubumarcos ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17d408531ac1a90d9d340e9886c50d5d601dc05adbeb15f7553e3e20fbb324ee
3
+ size 5806