Model save
Browse files
README.md
CHANGED
|
@@ -3,8 +3,6 @@ base_model: facebook/musicgen-small
|
|
| 3 |
library_name: peft
|
| 4 |
license: cc-by-nc-4.0
|
| 5 |
tags:
|
| 6 |
-
- text-to-audio
|
| 7 |
-
- ogbanugot/musicgen
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: musicgen-small-lora-afrobeats
|
|
@@ -17,10 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](None)
|
| 18 |
# musicgen-small-lora-afrobeats
|
| 19 |
|
| 20 |
-
This model is a fine-tuned version of [facebook/musicgen-small](https://huggingface.co/facebook/musicgen-small) on
|
| 21 |
-
It achieves the following results on the evaluation set:
|
| 22 |
-
- Loss: 6.8599
|
| 23 |
-
- Clap: 0.0514
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -47,25 +42,11 @@ The following hyperparameters were used during training:
|
|
| 47 |
- total_train_batch_size: 16
|
| 48 |
- optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-08
|
| 49 |
- lr_scheduler_type: linear
|
| 50 |
-
- num_epochs:
|
| 51 |
- mixed_precision_training: Native AMP
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|
| 55 |
-
| Training Loss | Epoch | Step | Validation Loss | Clap |
|
| 56 |
-
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 57 |
-
| 7.317 | 8.0 | 25 | 7.1357 | 0.0679 |
|
| 58 |
-
| 7.2825 | 16.0 | 50 | 7.0802 | 0.0755 |
|
| 59 |
-
| 7.1042 | 24.0 | 75 | 6.9064 | 0.0550 |
|
| 60 |
-
| 7.0441 | 32.0 | 100 | 6.8617 | 0.1064 |
|
| 61 |
-
| 6.8774 | 40.0 | 125 | 6.8261 | 0.0895 |
|
| 62 |
-
| 6.9014 | 48.0 | 150 | 6.8518 | 0.0798 |
|
| 63 |
-
| 7.0709 | 56.0 | 175 | 6.8745 | 0.0458 |
|
| 64 |
-
| 6.8413 | 64.0 | 200 | 6.8849 | 0.0387 |
|
| 65 |
-
| 6.9145 | 72.0 | 225 | 6.8753 | 0.0601 |
|
| 66 |
-
| 6.7884 | 80.0 | 250 | 6.8933 | 0.0301 |
|
| 67 |
-
| 6.827 | 88.0 | 275 | 6.8558 | 0.0403 |
|
| 68 |
-
| 6.9368 | 96.0 | 300 | 6.8599 | 0.0434 |
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|
|
|
|
| 3 |
library_name: peft
|
| 4 |
license: cc-by-nc-4.0
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: musicgen-small-lora-afrobeats
|
|
|
|
| 15 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](None)
|
| 16 |
# musicgen-small-lora-afrobeats
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [facebook/musicgen-small](https://huggingface.co/facebook/musicgen-small) on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 42 |
- total_train_batch_size: 16
|
| 43 |
- optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
+
- num_epochs: 4.0
|
| 46 |
- mixed_precision_training: Native AMP
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
|
| 51 |
|
| 52 |
### Framework versions
|