🚀 wav2vec2-large-960h-intent-classification-ori
該模型是基於facebook/wav2vec2-large-960h在None數據集上進行微調的版本。它在評估集上取得了以下效果：
📄 許可證
本項目採用Apache-2.0許可證。
🔧 技術細節
訓練超參數
訓練過程中使用了以下超參數：
- 學習率：3e-05
- 訓練批次大小：2
- 評估批次大小：2
- 隨機種子：42
- 梯度累積步數：4
- 總訓練批次大小：8
- 優化器：Adam（β1=0.9，β2=0.999，ε=1e-08）
- 學習率調度器類型：線性
- 學習率調度器熱身比例：0.1
- 訓練輪數：45
訓練結果
| 訓練損失 |
輪數 |
步數 |
驗證損失 |
準確率 |
| 2.1937 |
1.0 |
14 |
2.1715 |
0.3333 |
| 2.1764 |
2.0 |
28 |
2.1317 |
0.3333 |
| 2.1358 |
3.0 |
42 |
2.0794 |
0.3333 |
| 2.1227 |
4.0 |
56 |
2.0367 |
0.3333 |
| 2.1157 |
5.0 |
70 |
1.9845 |
0.3333 |
| 2.0776 |
6.0 |
84 |
1.9369 |
0.3333 |
| 2.0126 |
7.0 |
98 |
1.8021 |
0.375 |
| 1.88 |
8.0 |
112 |
1.6825 |
0.4167 |
| 1.8818 |
9.0 |
126 |
1.4638 |
0.5 |
| 1.7044 |
10.0 |
140 |
1.4188 |
0.5208 |
| 1.5442 |
11.0 |
154 |
1.3421 |
0.5625 |
| 1.5045 |
12.0 |
168 |
1.3971 |
0.5 |
| 1.3369 |
13.0 |
182 |
1.1602 |
0.5833 |
| 1.4017 |
14.0 |
196 |
1.3510 |
0.5417 |
| 1.2565 |
15.0 |
210 |
1.0978 |
0.5625 |
| 1.1056 |
16.0 |
224 |
1.0847 |
0.5833 |
| 1.2006 |
17.0 |
238 |
1.0262 |
0.625 |
| 0.9235 |
18.0 |
252 |
0.9532 |
0.7083 |
| 0.9528 |
19.0 |
266 |
1.0212 |
0.6042 |
| 0.8195 |
20.0 |
280 |
0.8442 |
0.7083 |
| 0.7518 |
21.0 |
294 |
0.8379 |
0.6875 |
| 0.6017 |
22.0 |
308 |
0.9422 |
0.7292 |
| 0.7697 |
23.0 |
322 |
0.7353 |
0.75 |
| 0.5367 |
24.0 |
336 |
0.8685 |
0.6875 |
| 0.5655 |
25.0 |
350 |
0.7440 |
0.7708 |
| 0.5116 |
26.0 |
364 |
0.7572 |
0.75 |
| 0.4297 |
27.0 |
378 |
0.7518 |
0.75 |
| 0.4928 |
28.0 |
392 |
0.5988 |
0.7917 |
| 0.4424 |
29.0 |
406 |
0.7240 |
0.75 |
| 0.3313 |
30.0 |
420 |
0.6173 |
0.7708 |
| 0.3854 |
31.0 |
434 |
0.7375 |
0.75 |
| 0.4131 |
32.0 |
448 |
0.7026 |
0.7708 |
| 0.2899 |
33.0 |
462 |
0.6516 |
0.7708 |
| 0.3644 |
34.0 |
476 |
0.6201 |
0.7917 |
| 0.2316 |
35.0 |
490 |
0.6111 |
0.7708 |
| 0.2589 |
36.0 |
504 |
0.5518 |
0.7917 |
| 0.3778 |
37.0 |
518 |
0.5512 |
0.7708 |
| 0.2426 |
38.0 |
532 |
0.5779 |
0.7917 |
| 0.304 |
39.0 |
546 |
0.7771 |
0.75 |
| 0.1833 |
40.0 |
560 |
0.5839 |
0.7708 |
| 0.1649 |
41.0 |
574 |
0.5699 |
0.7708 |
| 0.2529 |
42.0 |
588 |
0.6190 |
0.75 |
| 0.2121 |
43.0 |
602 |
0.5992 |
0.75 |
| 0.2736 |
44.0 |
616 |
0.6011 |
0.7917 |
| 0.2446 |
45.0 |
630 |
0.6013 |
0.7708 |
框架版本
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
PyTorch
Transformers 支持多種語言