Hugging Face's logo

guyPerry
/
audio-mood-classifier

Audio Classification
Transformers
Safetensors
English
audio-spectrogram-transformer
audio
ast
music
mood
Model card Files
xet
 Community
guyPerry commited on
Commit
7ae5115
·
verified ·
1 Parent(s): 67e3d2d

Upload fine-tuned audio mood classifier (AST)

Browse files
Files changed (5) hide show
  1. config.json +35 -0
  2. model.safetensors +3 -0
  3. scheduler.pt +3 -0
  4. trainer_state.json +215 -0
  5. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ASTForAudioClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "dtype": "float32",
7
+ "frequency_stride": 10,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2"
15
+ },
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "label2id": {
19
+ "LABEL_0": 0,
20
+ "LABEL_1": 1,
21
+ "LABEL_2": 2
22
+ },
23
+ "layer_norm_eps": 1e-12,
24
+ "max_length": 1024,
25
+ "model_type": "audio-spectrogram-transformer",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "num_mel_bins": 128,
29
+ "patch_size": 16,
30
+ "problem_type": "single_label_classification",
31
+ "qkv_bias": true,
32
+ "time_stride": 10,
33
+ "transformers_version": "5.13.1",
34
+ "use_cache": false
35
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40db84cfa01bac586ca8433339856705869dfe81992c046a97dd753a6f0278f6
3
+ size 344793116
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91e11dd4cff3282ec97b42fe6ec2025ed3b0d5d845391b86faee81d108d58104
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,215 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 372,
3
+ "best_metric": 0.39589157700538635,
4
+ "best_model_checkpoint": "F:\\Work_stuff\\VisualStudio_cursor\\audio_mood_classifier_hf\\models\\mood_classifier_2026-07-16_09-27\\checkpoint-372",
5
+ "epoch": 12.0,
6
+ "eval_steps": 500,
7
+ "global_step": 372,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 86.11,
15
+ "eval_gain_over_random": 52.78,
16
+ "eval_loss": 0.42865681648254395,
17
+ "eval_random_baseline": 33.33,
18
+ "eval_runtime": 46.4446,
19
+ "eval_samples_per_second": 4.651,
20
+ "eval_steps_per_second": 0.581,
21
+ "step": 31
22
+ },
23
+ {
24
+ "epoch": 1.6129032258064515,
25
+ "grad_norm": 5.450277328491211,
26
+ "learning_rate": 3.578494623655914e-07,
27
+ "loss": 0.4940774917602539,
28
+ "step": 50
29
+ },
30
+ {
31
+ "epoch": 2.0,
32
+ "eval_accuracy": 86.57,
33
+ "eval_gain_over_random": 53.24,
34
+ "eval_loss": 0.4212609827518463,
35
+ "eval_random_baseline": 33.33,
36
+ "eval_runtime": 46.6498,
37
+ "eval_samples_per_second": 4.63,
38
+ "eval_steps_per_second": 0.579,
39
+ "step": 62
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "eval_accuracy": 86.57,
44
+ "eval_gain_over_random": 53.24,
45
+ "eval_loss": 0.4187099039554596,
46
+ "eval_random_baseline": 33.33,
47
+ "eval_runtime": 46.4674,
48
+ "eval_samples_per_second": 4.648,
49
+ "eval_steps_per_second": 0.581,
50
+ "step": 93
51
+ },
52
+ {
53
+ "epoch": 3.225806451612903,
54
+ "grad_norm": 3.5786874294281006,
55
+ "learning_rate": 3.148387096774193e-07,
56
+ "loss": 0.4830916976928711,
57
+ "step": 100
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "eval_accuracy": 85.65,
62
+ "eval_gain_over_random": 52.31,
63
+ "eval_loss": 0.42973992228507996,
64
+ "eval_random_baseline": 33.33,
65
+ "eval_runtime": 46.5953,
66
+ "eval_samples_per_second": 4.636,
67
+ "eval_steps_per_second": 0.579,
68
+ "step": 124
69
+ },
70
+ {
71
+ "epoch": 4.838709677419355,
72
+ "grad_norm": 2.840916633605957,
73
+ "learning_rate": 2.718279569892473e-07,
74
+ "loss": 0.4838411331176758,
75
+ "step": 150
76
+ },
77
+ {
78
+ "epoch": 5.0,
79
+ "eval_accuracy": 83.8,
80
+ "eval_gain_over_random": 50.46,
81
+ "eval_loss": 0.4159768521785736,
82
+ "eval_random_baseline": 33.33,
83
+ "eval_runtime": 46.5148,
84
+ "eval_samples_per_second": 4.644,
85
+ "eval_steps_per_second": 0.58,
86
+ "step": 155
87
+ },
88
+ {
89
+ "epoch": 6.0,
90
+ "eval_accuracy": 85.19,
91
+ "eval_gain_over_random": 51.85,
92
+ "eval_loss": 0.4102042019367218,
93
+ "eval_random_baseline": 33.33,
94
+ "eval_runtime": 46.3609,
95
+ "eval_samples_per_second": 4.659,
96
+ "eval_steps_per_second": 0.582,
97
+ "step": 186
98
+ },
99
+ {
100
+ "epoch": 6.451612903225806,
101
+ "grad_norm": 3.7706756591796875,
102
+ "learning_rate": 2.2881720430107524e-07,
103
+ "loss": 0.47981719970703124,
104
+ "step": 200
105
+ },
106
+ {
107
+ "epoch": 7.0,
108
+ "eval_accuracy": 85.65,
109
+ "eval_gain_over_random": 52.31,
110
+ "eval_loss": 0.4190486669540405,
111
+ "eval_random_baseline": 33.33,
112
+ "eval_runtime": 43.6211,
113
+ "eval_samples_per_second": 4.952,
114
+ "eval_steps_per_second": 0.619,
115
+ "step": 217
116
+ },
117
+ {
118
+ "epoch": 8.0,
119
+ "eval_accuracy": 86.11,
120
+ "eval_gain_over_random": 52.78,
121
+ "eval_loss": 0.42142894864082336,
122
+ "eval_random_baseline": 33.33,
123
+ "eval_runtime": 52.8985,
124
+ "eval_samples_per_second": 4.083,
125
+ "eval_steps_per_second": 0.51,
126
+ "step": 248
127
+ },
128
+ {
129
+ "epoch": 8.064516129032258,
130
+ "grad_norm": 2.820573091506958,
131
+ "learning_rate": 1.8580645161290323e-07,
132
+ "loss": 0.4633090591430664,
133
+ "step": 250
134
+ },
135
+ {
136
+ "epoch": 9.0,
137
+ "eval_accuracy": 85.19,
138
+ "eval_gain_over_random": 51.85,
139
+ "eval_loss": 0.4239995777606964,
140
+ "eval_random_baseline": 33.33,
141
+ "eval_runtime": 54.2035,
142
+ "eval_samples_per_second": 3.985,
143
+ "eval_steps_per_second": 0.498,
144
+ "step": 279
145
+ },
146
+ {
147
+ "epoch": 9.67741935483871,
148
+ "grad_norm": 2.8738181591033936,
149
+ "learning_rate": 1.4279569892473117e-07,
150
+ "loss": 0.47099334716796876,
151
+ "step": 300
152
+ },
153
+ {
154
+ "epoch": 10.0,
155
+ "eval_accuracy": 85.19,
156
+ "eval_gain_over_random": 51.85,
157
+ "eval_loss": 0.4082905948162079,
158
+ "eval_random_baseline": 33.33,
159
+ "eval_runtime": 53.1952,
160
+ "eval_samples_per_second": 4.061,
161
+ "eval_steps_per_second": 0.508,
162
+ "step": 310
163
+ },
164
+ {
165
+ "epoch": 11.0,
166
+ "eval_accuracy": 87.96,
167
+ "eval_gain_over_random": 54.63,
168
+ "eval_loss": 0.4077238440513611,
169
+ "eval_random_baseline": 33.33,
170
+ "eval_runtime": 53.3942,
171
+ "eval_samples_per_second": 4.045,
172
+ "eval_steps_per_second": 0.506,
173
+ "step": 341
174
+ },
175
+ {
176
+ "epoch": 11.290322580645162,
177
+ "grad_norm": 3.939497232437134,
178
+ "learning_rate": 9.978494623655913e-08,
179
+ "loss": 0.47861801147460936,
180
+ "step": 350
181
+ },
182
+ {
183
+ "epoch": 12.0,
184
+ "eval_accuracy": 87.04,
185
+ "eval_gain_over_random": 53.7,
186
+ "eval_loss": 0.39589157700538635,
187
+ "eval_random_baseline": 33.33,
188
+ "eval_runtime": 53.13,
189
+ "eval_samples_per_second": 4.065,
190
+ "eval_steps_per_second": 0.508,
191
+ "step": 372
192
+ }
193
+ ],
194
+ "logging_steps": 50,
195
+ "max_steps": 465,
196
+ "num_input_tokens_seen": 0,
197
+ "num_train_epochs": 15,
198
+ "save_steps": 500,
199
+ "stateful_callbacks": {
200
+ "TrainerControl": {
201
+ "args": {
202
+ "should_epoch_stop": false,
203
+ "should_evaluate": false,
204
+ "should_log": false,
205
+ "should_save": true,
206
+ "should_training_stop": false
207
+ },
208
+ "attributes": {}
209
+ }
210
+ },
211
+ "total_flos": 8.052673225634611e+17,
212
+ "train_batch_size": 32,
213
+ "trial_name": null,
214
+ "trial_params": null
215
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16a0972a9ce454d8f2cdf556501a05522d7828ff4051516b59de9f4864ff67d2
3
+ size 4856