update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- audiofolder
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: wav2vec2-base-Drum_Kit_Sounds
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# wav2vec2-base-Drum_Kit_Sounds
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the audiofolder dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.0887
|
| 22 |
+
- Accuracy: 0.7812
|
| 23 |
+
- Weighted f1: 0.7692
|
| 24 |
+
- Micro f1: 0.7812
|
| 25 |
+
- Macro f1: 0.7845
|
| 26 |
+
- Weighted recall: 0.7812
|
| 27 |
+
- Micro recall: 0.7812
|
| 28 |
+
- Macro recall: 0.8187
|
| 29 |
+
- Weighted precision: 0.8717
|
| 30 |
+
- Micro precision: 0.7812
|
| 31 |
+
- Macro precision: 0.8534
|
| 32 |
+
|
| 33 |
+
## Model description
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Intended uses & limitations
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training and evaluation data
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training procedure
|
| 46 |
+
|
| 47 |
+
### Training hyperparameters
|
| 48 |
+
|
| 49 |
+
The following hyperparameters were used during training:
|
| 50 |
+
- learning_rate: 3e-05
|
| 51 |
+
- train_batch_size: 32
|
| 52 |
+
- eval_batch_size: 32
|
| 53 |
+
- seed: 42
|
| 54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
+
- lr_scheduler_type: linear
|
| 56 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 57 |
+
- num_epochs: 12
|
| 58 |
+
|
| 59 |
+
### Training results
|
| 60 |
+
|
| 61 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision |
|
| 62 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:--------:|:---------------:|:------------:|:------------:|:------------------:|:---------------:|:---------------:|
|
| 63 |
+
| 1.3743 | 1.0 | 4 | 1.3632 | 0.5625 | 0.5801 | 0.5625 | 0.5678 | 0.5625 | 0.5625 | 0.5670 | 0.6786 | 0.5625 | 0.6429 |
|
| 64 |
+
| 1.3074 | 2.0 | 8 | 1.3149 | 0.3438 | 0.2567 | 0.3438 | 0.2696 | 0.3438 | 0.3438 | 0.375 | 0.3067 | 0.3438 | 0.3148 |
|
| 65 |
+
| 1.2393 | 3.0 | 12 | 1.3121 | 0.2188 | 0.0785 | 0.2188 | 0.0897 | 0.2188 | 0.2188 | 0.25 | 0.0479 | 0.2188 | 0.0547 |
|
| 66 |
+
| 1.2317 | 4.0 | 16 | 1.3112 | 0.2812 | 0.1800 | 0.2812 | 0.2057 | 0.2812 | 0.2812 | 0.3214 | 0.2698 | 0.2812 | 0.3083 |
|
| 67 |
+
| 1.2107 | 5.0 | 20 | 1.2604 | 0.4375 | 0.3030 | 0.4375 | 0.3462 | 0.4375 | 0.4375 | 0.5 | 0.2552 | 0.4375 | 0.2917 |
|
| 68 |
+
| 1.1663 | 6.0 | 24 | 1.2112 | 0.4688 | 0.3896 | 0.4688 | 0.4310 | 0.4688 | 0.4688 | 0.5268 | 0.5041 | 0.4688 | 0.5404 |
|
| 69 |
+
| 1.1247 | 7.0 | 28 | 1.1746 | 0.5938 | 0.5143 | 0.5938 | 0.5603 | 0.5938 | 0.5938 | 0.6562 | 0.5220 | 0.5938 | 0.5609 |
|
| 70 |
+
| 1.0856 | 8.0 | 32 | 1.1434 | 0.5938 | 0.5143 | 0.5938 | 0.5603 | 0.5938 | 0.5938 | 0.6562 | 0.5220 | 0.5938 | 0.5609 |
|
| 71 |
+
| 1.0601 | 9.0 | 36 | 1.1417 | 0.6562 | 0.6029 | 0.6562 | 0.6389 | 0.6562 | 0.6562 | 0.7125 | 0.8440 | 0.6562 | 0.8217 |
|
| 72 |
+
| 1.0375 | 10.0 | 40 | 1.1227 | 0.6875 | 0.6582 | 0.6875 | 0.6831 | 0.6875 | 0.6875 | 0.7330 | 0.8457 | 0.6875 | 0.8237 |
|
| 73 |
+
| 1.0168 | 11.0 | 44 | 1.1065 | 0.7812 | 0.7692 | 0.7812 | 0.7845 | 0.7812 | 0.7812 | 0.8187 | 0.8717 | 0.7812 | 0.8534 |
|
| 74 |
+
| 1.0093 | 12.0 | 48 | 1.0887 | 0.7812 | 0.7692 | 0.7812 | 0.7845 | 0.7812 | 0.7812 | 0.8187 | 0.8717 | 0.7812 | 0.8534 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.25.1
|
| 80 |
+
- Pytorch 1.12.1
|
| 81 |
+
- Datasets 2.8.0
|
| 82 |
+
- Tokenizers 0.12.1
|