<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wav2vec2-tiny-1-cnn-fp16-demo
This model is a fine-tuned version of yenpolin/wav2vec2-tiny-1-cnn on the DNA_R9.4.1 - NA dataset. It achieves the following results on the evaluation set:
- Loss: 0.2063
- Mean Acc: 2.4426
- Median Acc: 0.0
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 100
- eval_batch_size: 200
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 100.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Mean Acc | Median Acc |
---|---|---|---|---|---|
0.5146 | 1.0 | 1000 | 0.3673 | 0.0 | 0.0 |
0.3785 | 2.0 | 2000 | 0.3447 | 0.0 | 0.0 |
0.364 | 3.0 | 3000 | 0.3317 | 0.0 | 0.0 |
0.3523 | 4.0 | 4000 | 0.3249 | 0.0027 | 0.0 |
0.3447 | 5.0 | 5000 | 0.3181 | 0.0027 | 0.0 |
0.3386 | 6.0 | 6000 | 0.3123 | 0.0027 | 0.0 |
0.3334 | 7.0 | 7000 | 0.3085 | 0.0027 | 0.0 |
0.3294 | 8.0 | 8000 | 0.3046 | 0.0 | 0.0 |
0.326 | 9.0 | 9000 | 0.3027 | 0.0 | 0.0 |
0.323 | 10.0 | 10000 | 0.2978 | 0.0 | 0.0 |
0.32 | 11.0 | 11000 | 0.2963 | 0.0027 | 0.0 |
0.3061 | 12.0 | 12000 | 0.2616 | 0.0465 | 0.0 |
0.2818 | 13.0 | 13000 | 0.2575 | 0.0486 | 0.0 |
0.2768 | 14.0 | 14000 | 0.2517 | 0.1075 | 0.0 |
0.2731 | 15.0 | 15000 | 0.2492 | 0.0397 | 0.0 |
0.2697 | 16.0 | 16000 | 0.2448 | 0.0648 | 0.0 |
0.2658 | 17.0 | 17000 | 0.2444 | 0.1171 | 0.0 |
0.2619 | 18.0 | 18000 | 0.2385 | 0.1305 | 0.0 |
0.2585 | 19.0 | 19000 | 0.2389 | 0.0960 | 0.0 |
0.2546 | 20.0 | 20000 | 0.2326 | 0.0909 | 0.0 |
0.2505 | 21.0 | 21000 | 0.2287 | 0.0635 | 0.0 |
0.2462 | 22.0 | 22000 | 0.2240 | 0.1070 | 0.0 |
0.2408 | 23.0 | 23000 | 0.2190 | 0.1429 | 0.0 |
0.2334 | 24.0 | 24000 | 0.2109 | 0.7212 | 0.0 |
0.225 | 25.0 | 25000 | 0.2027 | 0.7305 | 0.0 |
0.2178 | 26.0 | 26000 | 0.1980 | 0.8186 | 0.0 |
0.2115 | 27.0 | 27000 | 0.1937 | 0.9743 | 0.0 |
0.2065 | 28.0 | 28000 | 0.1892 | 0.8266 | 0.0 |
0.2026 | 29.0 | 29000 | 0.1890 | 0.1615 | 0.0 |
0.1987 | 30.0 | 30000 | 0.1836 | 1.0021 | 0.0 |
0.1953 | 31.0 | 31000 | 0.1830 | 0.8009 | 0.0 |
0.1921 | 32.0 | 32000 | 0.1821 | 1.2837 | 0.0 |
0.1893 | 33.0 | 33000 | 0.1819 | 0.5987 | 0.0 |
0.1865 | 34.0 | 34000 | 0.1835 | 0.9360 | 0.0 |
0.1835 | 35.0 | 35000 | 0.1796 | 1.3452 | 0.0 |
0.1808 | 36.0 | 36000 | 0.1816 | 1.4669 | 0.0 |
0.1779 | 37.0 | 37000 | 0.1806 | 2.4269 | 0.0 |
0.1755 | 38.0 | 38000 | 0.1787 | 0.7843 | 0.0 |
0.1726 | 39.0 | 39000 | 0.1807 | 1.8650 | 0.0 |
0.1699 | 40.0 | 40000 | 0.1811 | 2.1893 | 0.0 |
0.167 | 41.0 | 41000 | 0.1799 | 1.7285 | 0.0 |
0.1644 | 42.0 | 42000 | 0.1792 | 1.5862 | 0.0 |
0.1617 | 43.0 | 43000 | 0.1785 | 1.5165 | 0.0 |
0.159 | 44.0 | 44000 | 0.1806 | 1.1542 | 0.0 |
0.1563 | 45.0 | 45000 | 0.1804 | 1.8334 | 0.0 |
0.1539 | 46.0 | 46000 | 0.1830 | 2.1450 | 0.0 |
0.1515 | 47.0 | 47000 | 0.1835 | 2.2905 | 0.0 |
0.1489 | 48.0 | 48000 | 0.1821 | 2.4879 | 0.0 |
0.1465 | 49.0 | 49000 | 0.1806 | 1.6113 | 0.0 |
0.1441 | 50.0 | 50000 | 0.1857 | 1.7132 | 0.0 |
0.1418 | 51.0 | 51000 | 0.1847 | 1.2830 | 0.0 |
0.1394 | 52.0 | 52000 | 0.1862 | 1.8548 | 0.0 |
0.137 | 53.0 | 53000 | 0.1815 | 2.1585 | 0.0 |
0.1345 | 54.0 | 54000 | 0.1896 | 1.1918 | 0.0 |
0.1325 | 55.0 | 55000 | 0.1892 | 1.5508 | 0.0 |
0.1304 | 56.0 | 56000 | 0.1879 | 1.5180 | 0.0 |
0.1282 | 57.0 | 57000 | 0.1868 | 0.8048 | 0.0 |
0.126 | 58.0 | 58000 | 0.1906 | 1.6521 | 0.0 |
0.1243 | 59.0 | 59000 | 0.1879 | 1.3255 | 0.0 |
0.1222 | 60.0 | 60000 | 0.1884 | 1.2970 | 0.0 |
0.1202 | 61.0 | 61000 | 0.1905 | 1.5370 | 0.0 |
0.1184 | 62.0 | 62000 | 0.1936 | 1.7408 | 0.0 |
0.1167 | 63.0 | 63000 | 0.1922 | 1.5556 | 0.0 |
0.1149 | 64.0 | 64000 | 0.1960 | 1.5176 | 0.0 |
0.1133 | 65.0 | 65000 | 0.1966 | 1.8577 | 0.0 |
0.1117 | 66.0 | 66000 | 0.1942 | 2.1886 | 0.0 |
0.1102 | 67.0 | 67000 | 0.1961 | 1.4547 | 0.0 |
0.1087 | 68.0 | 68000 | 0.1965 | 1.5482 | 0.0 |
0.1072 | 69.0 | 69000 | 0.1965 | 1.5644 | 0.0 |
0.1058 | 70.0 | 70000 | 0.1982 | 1.5436 | 0.0 |
0.1045 | 71.0 | 71000 | 0.1956 | 2.2319 | 0.0 |
0.1033 | 72.0 | 72000 | 0.2037 | 2.5393 | 0.0 |
0.1019 | 73.0 | 73000 | 0.1977 | 1.9089 | 0.0 |
0.1008 | 74.0 | 74000 | 0.1983 | 1.6957 | 0.0 |
0.0997 | 75.0 | 75000 | 0.1980 | 3.0164 | 0.0 |
0.0986 | 76.0 | 76000 | 0.2004 | 1.8043 | 0.0 |
0.0976 | 77.0 | 77000 | 0.2025 | 1.8234 | 0.0 |
0.0966 | 78.0 | 78000 | 0.2006 | 2.1719 | 0.0 |
0.0957 | 79.0 | 79000 | 0.1991 | 1.8594 | 0.0 |
0.0948 | 80.0 | 80000 | 0.2037 | 1.9660 | 0.0 |
0.094 | 81.0 | 81000 | 0.2000 | 2.2630 | 0.0 |
0.0932 | 82.0 | 82000 | 0.2068 | 2.3164 | 0.0 |
0.0925 | 83.0 | 83000 | 0.2004 | 2.0110 | 0.0 |
0.0918 | 84.0 | 84000 | 0.2034 | 2.4419 | 0.0 |
0.0912 | 85.0 | 85000 | 0.2031 | 2.0823 | 0.0 |
0.0906 | 86.0 | 86000 | 0.2039 | 2.3955 | 0.0 |
0.0901 | 87.0 | 87000 | 0.2042 | 2.4907 | 0.0 |
0.0897 | 88.0 | 88000 | 0.2052 | 2.3399 | 0.0 |
0.0892 | 89.0 | 89000 | 0.2050 | 2.3214 | 0.0 |
0.0888 | 90.0 | 90000 | 0.2052 | 2.1339 | 0.0 |
0.0884 | 91.0 | 91000 | 0.2050 | 2.3432 | 0.0 |
0.0881 | 92.0 | 92000 | 0.2052 | 2.3161 | 0.0 |
0.0878 | 93.0 | 93000 | 0.2050 | 2.4258 | 0.0 |
0.0876 | 94.0 | 94000 | 0.2051 | 2.2742 | 0.0 |
0.0874 | 95.0 | 95000 | 0.2065 | 2.3400 | 0.0 |
0.0872 | 96.0 | 96000 | 0.2063 | 2.4099 | 0.0 |
0.0871 | 97.0 | 97000 | 0.2060 | 2.4249 | 0.0 |
0.087 | 98.0 | 98000 | 0.2064 | 2.4826 | 0.0 |
0.0869 | 99.0 | 99000 | 0.2064 | 2.4447 | 0.0 |
0.0869 | 100.0 | 100000 | 0.2063 | 2.4426 | 0.0 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.12.1
- Datasets 2.9.0
- Tokenizers 0.13.2