Model save

README.md

@@ -3,8 +3,6 @@ library_name: peft
 license: cc-by-nc-4.0
 base_model: facebook/musicgen-melody
 tags:
-- text-to-audio
-- ylacombe/tiny-punk
 - generated_from_trainer
 model-index:
 - name: musicgen-melody-lora-punk
@@ -16,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # musicgen-melody-lora-punk
 
-This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the YLACOMBE/TINY-PUNK - DEFAULT dataset.
+This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 5.4128
-- Clap: -0.0280
+- Loss: 6.4997
+- Clap: 0.1150
 
 ## Model description
 
@@ -39,11 +37,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 2
+- train_batch_size: 1
 - eval_batch_size: 1
 - seed: 456
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 16
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 4
 - optimizer: Use adamw_torch with betas=(0.9,0.99) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 10.0
@@ -51,6 +49,9 @@ The following hyperparameters were used during training:
 
 ### Training results
 
+| Training Loss | Epoch  | Step | Validation Loss | Clap   |
+|:-------------:|:------:|:----:|:---------------:|:------:|
+| 6.9552        | 5.6061 | 50   | 6.4997          | 0.1150 |
 
 
 ### Framework versions