--- tags: - automatic-speech-recognition - gary109/AI_Light_Dance - generated_from_trainer datasets: - ai_light_dance metrics: - wer model-index: - name: ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v3 results: [] --- # ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v3 This model is a fine-tuned version of [gary109/ai-light-dance_drums_pretrain_wav2vec2-base-new](https://huggingface.co/gary109/ai-light-dance_drums_pretrain_wav2vec2-base-new) on the GARY109/AI_LIGHT_DANCE - ONSET-IDMT-SMT-DRUMS-V2+MDBDRUMS dataset. It achieves the following results on the evaluation set: - Loss: 1.0118 - Wer: 0.5852 ## 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: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 64 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 20 - num_epochs: 100.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | No log | 0.7 | 2 | 65.6850 | 0.9201 | | No log | 1.7 | 4 | 65.3150 | 0.9166 | | No log | 2.7 | 6 | 63.4014 | 0.8987 | | No log | 3.7 | 8 | 61.8469 | 0.8701 | | 20.497 | 4.7 | 10 | 57.4674 | 0.8558 | | 20.497 | 5.7 | 12 | 51.3715 | 0.9356 | | 20.497 | 6.7 | 14 | 43.4658 | 0.9869 | | 20.497 | 7.7 | 16 | 33.7960 | 1.0 | | 20.497 | 8.7 | 18 | 22.8680 | 1.0 | | 13.2039 | 9.7 | 20 | 12.0107 | 1.0 | | 13.2039 | 10.7 | 22 | 4.6754 | 1.0 | | 13.2039 | 11.7 | 24 | 2.5355 | 1.0 | | 13.2039 | 12.7 | 26 | 2.2341 | 1.0 | | 13.2039 | 13.7 | 28 | 2.8693 | 1.0 | | 3.0397 | 14.7 | 30 | 3.5083 | 1.0 | | 3.0397 | 15.7 | 32 | 3.1664 | 1.0 | | 3.0397 | 16.7 | 34 | 2.6172 | 1.0 | | 3.0397 | 17.7 | 36 | 2.2690 | 1.0 | | 3.0397 | 18.7 | 38 | 2.3809 | 1.0 | | 2.133 | 19.7 | 40 | 2.2537 | 1.0 | | 2.133 | 20.7 | 42 | 1.9852 | 1.0 | | 2.133 | 21.7 | 44 | 2.0724 | 1.0 | | 2.133 | 22.7 | 46 | 2.0076 | 1.0 | | 2.133 | 23.7 | 48 | 1.9461 | 1.0 | | 1.8964 | 24.7 | 50 | 2.0851 | 1.0 | | 1.8964 | 25.7 | 52 | 2.0501 | 1.0 | | 1.8964 | 26.7 | 54 | 1.8149 | 1.0 | | 1.8964 | 27.7 | 56 | 1.8059 | 1.0 | | 1.8964 | 28.7 | 58 | 1.9727 | 1.0 | | 1.8599 | 29.7 | 60 | 1.7996 | 1.0 | | 1.8599 | 30.7 | 62 | 1.6452 | 1.0 | | 1.8599 | 31.7 | 64 | 1.8071 | 1.0 | | 1.8599 | 32.7 | 66 | 1.7773 | 1.0 | | 1.8599 | 33.7 | 68 | 1.6234 | 1.0 | | 1.7166 | 34.7 | 70 | 1.6564 | 1.0 | | 1.7166 | 35.7 | 72 | 1.6550 | 1.0 | | 1.7166 | 36.7 | 74 | 1.5652 | 1.0 | | 1.7166 | 37.7 | 76 | 1.5213 | 0.9952 | | 1.7166 | 38.7 | 78 | 1.4795 | 0.9857 | | 1.4764 | 39.7 | 80 | 1.4402 | 0.9869 | | 1.4764 | 40.7 | 82 | 1.4425 | 0.9416 | | 1.4764 | 41.7 | 84 | 1.4473 | 0.8462 | | 1.4764 | 42.7 | 86 | 1.3786 | 0.8999 | | 1.4764 | 43.7 | 88 | 1.4610 | 0.8486 | | 1.3782 | 44.7 | 90 | 1.6983 | 0.8498 | | 1.3782 | 45.7 | 92 | 1.5572 | 0.9130 | | 1.3782 | 46.7 | 94 | 1.6037 | 0.8129 | | 1.3782 | 47.7 | 96 | 1.6875 | 0.7151 | | 1.3782 | 48.7 | 98 | 1.6262 | 0.7783 | | 1.3406 | 49.7 | 100 | 1.6009 | 0.7306 | | 1.3406 | 50.7 | 102 | 1.5745 | 0.6758 | | 1.3406 | 51.7 | 104 | 1.5348 | 0.6806 | | 1.3406 | 52.7 | 106 | 1.5455 | 0.6591 | | 1.3406 | 53.7 | 108 | 1.4916 | 0.6293 | | 1.1939 | 54.7 | 110 | 1.3992 | 0.6317 | | 1.1939 | 55.7 | 112 | 1.3632 | 0.6079 | | 1.1939 | 56.7 | 114 | 1.2971 | 0.6126 | | 1.1939 | 57.7 | 116 | 1.2397 | 0.6257 | | 1.1939 | 58.7 | 118 | 1.2213 | 0.6114 | | 1.2298 | 59.7 | 120 | 1.2703 | 0.6007 | | 1.2298 | 60.7 | 122 | 1.3285 | 0.5936 | | 1.2298 | 61.7 | 124 | 1.4112 | 0.5995 | | 1.2298 | 62.7 | 126 | 1.4664 | 0.5995 | | 1.2298 | 63.7 | 128 | 1.5034 | 0.6043 | | 1.1321 | 64.7 | 130 | 1.4718 | 0.6126 | | 1.1321 | 65.7 | 132 | 1.4657 | 0.6293 | | 1.1321 | 66.7 | 134 | 1.4940 | 0.6007 | | 1.1321 | 67.7 | 136 | 1.5151 | 0.5900 | | 1.1321 | 68.7 | 138 | 1.4332 | 0.6019 | | 1.068 | 69.7 | 140 | 1.3177 | 0.6138 | | 1.068 | 70.7 | 142 | 1.2636 | 0.6138 | | 1.068 | 71.7 | 144 | 1.2209 | 0.6007 | | 1.068 | 72.7 | 146 | 1.1464 | 0.6019 | | 1.068 | 73.7 | 148 | 1.0894 | 0.6246 | | 1.0462 | 74.7 | 150 | 1.0838 | 0.6246 | | 1.0462 | 75.7 | 152 | 1.0775 | 0.6222 | | 1.0462 | 76.7 | 154 | 1.0625 | 0.6114 | | 1.0462 | 77.7 | 156 | 1.0521 | 0.6043 | | 1.0462 | 78.7 | 158 | 1.0450 | 0.5995 | | 1.0199 | 79.7 | 160 | 1.0607 | 0.5948 | | 1.0199 | 80.7 | 162 | 1.0569 | 0.5924 | | 1.0199 | 81.7 | 164 | 1.0501 | 0.5912 | | 1.0199 | 82.7 | 166 | 1.0511 | 0.5888 | | 1.0199 | 83.7 | 168 | 1.0690 | 0.5828 | | 0.9651 | 84.7 | 170 | 1.0685 | 0.5805 | | 0.9651 | 85.7 | 172 | 1.0488 | 0.5864 | | 0.9651 | 86.7 | 174 | 1.0316 | 0.5900 | | 0.9651 | 87.7 | 176 | 1.0346 | 0.5900 | | 0.9651 | 88.7 | 178 | 1.0490 | 0.5828 | | 0.9867 | 89.7 | 180 | 1.0467 | 0.5828 | | 0.9867 | 90.7 | 182 | 1.0315 | 0.5888 | | 0.9867 | 91.7 | 184 | 1.0206 | 0.5912 | | 0.9867 | 92.7 | 186 | 1.0165 | 0.5924 | | 0.9867 | 93.7 | 188 | 1.0147 | 0.5900 | | 0.9332 | 94.7 | 190 | 1.0174 | 0.5876 | | 0.9332 | 95.7 | 192 | 1.0195 | 0.5852 | | 0.9332 | 96.7 | 194 | 1.0195 | 0.5840 | | 0.9332 | 97.7 | 196 | 1.0152 | 0.5840 | | 0.9332 | 98.7 | 198 | 1.0118 | 0.5852 | | 0.9299 | 99.7 | 200 | 1.0128 | 0.5852 | ### Framework versions - Transformers 4.25.0.dev0 - Pytorch 1.8.1+cu111 - Datasets 2.7.1.dev0 - Tokenizers 0.13.2