update model card README.md
Browse files
README.md
CHANGED
|
@@ -5,9 +5,24 @@ tags:
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- marsyas/gtzan
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: ast-finetuned-gtzan
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -17,13 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [MIT/ast-finetuned-audioset-10-10-0.4593](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593) on the GTZAN dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
-
|
| 21 |
-
-
|
| 22 |
-
- eval_runtime: 12.0837
|
| 23 |
-
- eval_samples_per_second: 8.276
|
| 24 |
-
- eval_steps_per_second: 2.069
|
| 25 |
-
- epoch: 4.0
|
| 26 |
-
- step: 900
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
@@ -51,6 +61,32 @@ The following hyperparameters were used during training:
|
|
| 51 |
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
- num_epochs: 20
|
| 53 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
### Framework versions
|
| 55 |
|
| 56 |
- Transformers 4.32.0.dev0
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- marsyas/gtzan
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: ast-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: 0.93
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [MIT/ast-finetuned-audioset-10-10-0.4593](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593) on the GTZAN dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.4436
|
| 36 |
+
- Accuracy: 0.93
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 61 |
- lr_scheduler_warmup_ratio: 0.1
|
| 62 |
- num_epochs: 20
|
| 63 |
|
| 64 |
+
### Training results
|
| 65 |
+
|
| 66 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 67 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 68 |
+
| 0.0001 | 1.0 | 225 | 0.5546 | 0.89 |
|
| 69 |
+
| 1.204 | 2.0 | 450 | 0.9484 | 0.81 |
|
| 70 |
+
| 0.4719 | 3.0 | 675 | 0.7417 | 0.85 |
|
| 71 |
+
| 0.0132 | 4.0 | 900 | 0.7101 | 0.9 |
|
| 72 |
+
| 0.0527 | 5.0 | 1125 | 0.8170 | 0.86 |
|
| 73 |
+
| 0.0 | 6.0 | 1350 | 0.6406 | 0.93 |
|
| 74 |
+
| 0.3099 | 7.0 | 1575 | 0.8426 | 0.84 |
|
| 75 |
+
| 0.0 | 8.0 | 1800 | 0.9173 | 0.89 |
|
| 76 |
+
| 0.0 | 9.0 | 2025 | 0.7142 | 0.9 |
|
| 77 |
+
| 0.0602 | 10.0 | 2250 | 0.4718 | 0.92 |
|
| 78 |
+
| 0.0003 | 11.0 | 2475 | 0.9860 | 0.9 |
|
| 79 |
+
| 0.0001 | 12.0 | 2700 | 0.5918 | 0.91 |
|
| 80 |
+
| 0.0 | 13.0 | 2925 | 0.4886 | 0.92 |
|
| 81 |
+
| 0.0 | 14.0 | 3150 | 0.4562 | 0.93 |
|
| 82 |
+
| 0.0 | 15.0 | 3375 | 0.4360 | 0.94 |
|
| 83 |
+
| 0.0 | 16.0 | 3600 | 0.4433 | 0.94 |
|
| 84 |
+
| 0.0 | 17.0 | 3825 | 0.4454 | 0.94 |
|
| 85 |
+
| 0.0 | 18.0 | 4050 | 0.4454 | 0.94 |
|
| 86 |
+
| 0.0 | 19.0 | 4275 | 0.4434 | 0.93 |
|
| 87 |
+
| 0.0 | 20.0 | 4500 | 0.4436 | 0.93 |
|
| 88 |
+
|
| 89 |
+
|
| 90 |
### Framework versions
|
| 91 |
|
| 92 |
- Transformers 4.32.0.dev0
|