Maksimkrug
commited on
Commit
•
f27115f
1
Parent(s):
fc65c6d
End of training
Browse files- README.md +84 -0
- config.json +176 -0
- model.safetensors +3 -0
- preprocessor_config.json +14 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: openai/whisper-small
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- marsyas/gtzan
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: whisper-small-finetuned-gtzan
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Audio Classification
|
15 |
+
type: audio-classification
|
16 |
+
dataset:
|
17 |
+
name: GTZAN
|
18 |
+
type: marsyas/gtzan
|
19 |
+
config: all
|
20 |
+
split: train
|
21 |
+
args: all
|
22 |
+
metrics:
|
23 |
+
- name: Accuracy
|
24 |
+
type: accuracy
|
25 |
+
value: 0.92
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# whisper-small-finetuned-gtzan
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the GTZAN dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.3487
|
36 |
+
- Accuracy: 0.92
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 4e-05
|
56 |
+
- train_batch_size: 2
|
57 |
+
- eval_batch_size: 2
|
58 |
+
- seed: 42
|
59 |
+
- gradient_accumulation_steps: 4
|
60 |
+
- total_train_batch_size: 8
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- lr_scheduler_warmup_ratio: 0.1
|
64 |
+
- num_epochs: 7
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
69 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
70 |
+
| 1.4671 | 1.0 | 112 | 0.9088 | 0.74 |
|
71 |
+
| 0.6277 | 2.0 | 225 | 0.6298 | 0.8 |
|
72 |
+
| 0.446 | 3.0 | 337 | 0.4495 | 0.86 |
|
73 |
+
| 0.19 | 4.0 | 450 | 0.3389 | 0.91 |
|
74 |
+
| 0.2302 | 5.0 | 562 | 0.2954 | 0.93 |
|
75 |
+
| 0.0031 | 6.0 | 675 | 0.3213 | 0.94 |
|
76 |
+
| 0.0021 | 6.97 | 784 | 0.3487 | 0.92 |
|
77 |
+
|
78 |
+
|
79 |
+
### Framework versions
|
80 |
+
|
81 |
+
- Transformers 4.35.0
|
82 |
+
- Pytorch 2.1.0+cu121
|
83 |
+
- Datasets 2.14.6
|
84 |
+
- Tokenizers 0.14.1
|
config.json
ADDED
@@ -0,0 +1,176 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"apply_spec_augment": false,
|
6 |
+
"architectures": [
|
7 |
+
"WhisperForAudioClassification"
|
8 |
+
],
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"begin_suppress_tokens": [
|
11 |
+
220,
|
12 |
+
50257
|
13 |
+
],
|
14 |
+
"bos_token_id": 50257,
|
15 |
+
"classifier_proj_size": 256,
|
16 |
+
"d_model": 768,
|
17 |
+
"decoder_attention_heads": 12,
|
18 |
+
"decoder_ffn_dim": 3072,
|
19 |
+
"decoder_layerdrop": 0.0,
|
20 |
+
"decoder_layers": 12,
|
21 |
+
"decoder_start_token_id": 50258,
|
22 |
+
"dropout": 0.0,
|
23 |
+
"encoder_attention_heads": 12,
|
24 |
+
"encoder_ffn_dim": 3072,
|
25 |
+
"encoder_layerdrop": 0.0,
|
26 |
+
"encoder_layers": 12,
|
27 |
+
"eos_token_id": 50257,
|
28 |
+
"forced_decoder_ids": [
|
29 |
+
[
|
30 |
+
1,
|
31 |
+
50259
|
32 |
+
],
|
33 |
+
[
|
34 |
+
2,
|
35 |
+
50359
|
36 |
+
],
|
37 |
+
[
|
38 |
+
3,
|
39 |
+
50363
|
40 |
+
]
|
41 |
+
],
|
42 |
+
"id2label": {
|
43 |
+
"0": "blues",
|
44 |
+
"1": "classical",
|
45 |
+
"2": "country",
|
46 |
+
"3": "disco",
|
47 |
+
"4": "hiphop",
|
48 |
+
"5": "jazz",
|
49 |
+
"6": "metal",
|
50 |
+
"7": "pop",
|
51 |
+
"8": "reggae",
|
52 |
+
"9": "rock"
|
53 |
+
},
|
54 |
+
"init_std": 0.02,
|
55 |
+
"is_encoder_decoder": true,
|
56 |
+
"label2id": {
|
57 |
+
"blues": "0",
|
58 |
+
"classical": "1",
|
59 |
+
"country": "2",
|
60 |
+
"disco": "3",
|
61 |
+
"hiphop": "4",
|
62 |
+
"jazz": "5",
|
63 |
+
"metal": "6",
|
64 |
+
"pop": "7",
|
65 |
+
"reggae": "8",
|
66 |
+
"rock": "9"
|
67 |
+
},
|
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_prob": 0.05,
|
74 |
+
"max_length": 448,
|
75 |
+
"max_source_positions": 1500,
|
76 |
+
"max_target_positions": 448,
|
77 |
+
"median_filter_width": 7,
|
78 |
+
"model_type": "whisper",
|
79 |
+
"num_hidden_layers": 12,
|
80 |
+
"num_mel_bins": 80,
|
81 |
+
"pad_token_id": 50257,
|
82 |
+
"scale_embedding": false,
|
83 |
+
"suppress_tokens": [
|
84 |
+
1,
|
85 |
+
2,
|
86 |
+
7,
|
87 |
+
8,
|
88 |
+
9,
|
89 |
+
10,
|
90 |
+
14,
|
91 |
+
25,
|
92 |
+
26,
|
93 |
+
27,
|
94 |
+
28,
|
95 |
+
29,
|
96 |
+
31,
|
97 |
+
58,
|
98 |
+
59,
|
99 |
+
60,
|
100 |
+
61,
|
101 |
+
62,
|
102 |
+
63,
|
103 |
+
90,
|
104 |
+
91,
|
105 |
+
92,
|
106 |
+
93,
|
107 |
+
359,
|
108 |
+
503,
|
109 |
+
522,
|
110 |
+
542,
|
111 |
+
873,
|
112 |
+
893,
|
113 |
+
902,
|
114 |
+
918,
|
115 |
+
922,
|
116 |
+
931,
|
117 |
+
1350,
|
118 |
+
1853,
|
119 |
+
1982,
|
120 |
+
2460,
|
121 |
+
2627,
|
122 |
+
3246,
|
123 |
+
3253,
|
124 |
+
3268,
|
125 |
+
3536,
|
126 |
+
3846,
|
127 |
+
3961,
|
128 |
+
4183,
|
129 |
+
4667,
|
130 |
+
6585,
|
131 |
+
6647,
|
132 |
+
7273,
|
133 |
+
9061,
|
134 |
+
9383,
|
135 |
+
10428,
|
136 |
+
10929,
|
137 |
+
11938,
|
138 |
+
12033,
|
139 |
+
12331,
|
140 |
+
12562,
|
141 |
+
13793,
|
142 |
+
14157,
|
143 |
+
14635,
|
144 |
+
15265,
|
145 |
+
15618,
|
146 |
+
16553,
|
147 |
+
16604,
|
148 |
+
18362,
|
149 |
+
18956,
|
150 |
+
20075,
|
151 |
+
21675,
|
152 |
+
22520,
|
153 |
+
26130,
|
154 |
+
26161,
|
155 |
+
26435,
|
156 |
+
28279,
|
157 |
+
29464,
|
158 |
+
31650,
|
159 |
+
32302,
|
160 |
+
32470,
|
161 |
+
36865,
|
162 |
+
42863,
|
163 |
+
47425,
|
164 |
+
49870,
|
165 |
+
50254,
|
166 |
+
50258,
|
167 |
+
50360,
|
168 |
+
50361,
|
169 |
+
50362
|
170 |
+
],
|
171 |
+
"torch_dtype": "float32",
|
172 |
+
"transformers_version": "4.35.0",
|
173 |
+
"use_cache": true,
|
174 |
+
"use_weighted_layer_sum": false,
|
175 |
+
"vocab_size": 51865
|
176 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65a1cfd0fda3adf0cfcdd379e8a01854e91b5158d28179c7eafe984e113c51a5
|
3 |
+
size 353434592
|
preprocessor_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length": 30,
|
3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
4 |
+
"feature_size": 80,
|
5 |
+
"hop_length": 160,
|
6 |
+
"n_fft": 400,
|
7 |
+
"n_samples": 480000,
|
8 |
+
"nb_max_frames": 3000,
|
9 |
+
"padding_side": "right",
|
10 |
+
"padding_value": 0.0,
|
11 |
+
"processor_class": "WhisperProcessor",
|
12 |
+
"return_attention_mask": false,
|
13 |
+
"sampling_rate": 16000
|
14 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:483eb2a1b1d4e4ae5b9cb2c6896228b997822729af693db1fe027ed26f05f3db
|
3 |
+
size 4600
|