End of training

README.md

@@ -3,6 +3,8 @@ base_model: facebook/musicgen-melody
 library_name: peft
 license: cc-by-nc-4.0
 tags:
+- text-to-audio
+- Punk
 - generated_from_trainer
 model-index:
 - name: Generated
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Generated
 
-This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on an unknown dataset.
+This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the ylacombe/tiny-punk dataset.
 
 ## Model description
 

trainer_state.json

@@ -1,109 +1,67 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.84,
+  "epoch": 3.5555555555555554,
   "eval_steps": 500,
-  "global_step": 24,
+  "global_step": 12,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.32,
-      "grad_norm": 1.0916129350662231,
-      "learning_rate": 0.00018333333333333334,
-      "loss": 9.4256,
-      "step": 2
-    },
-    {
-      "epoch": 0.64,
-      "grad_norm": 1.684420108795166,
+      "epoch": 0.5925925925925926,
+      "grad_norm": 1.2005751132965088,
       "learning_rate": 0.0001666666666666667,
-      "loss": 9.0194,
-      "step": 4
-    },
-    {
-      "epoch": 0.96,
-      "grad_norm": 1.849820852279663,
-      "learning_rate": 0.00015000000000000001,
-      "loss": 8.5285,
-      "step": 6
+      "loss": 9.4962,
+      "step": 2
     },
     {
-      "epoch": 1.28,
-      "grad_norm": 2.3102447986602783,
+      "epoch": 1.1851851851851851,
+      "grad_norm": 1.47324538230896,
       "learning_rate": 0.00013333333333333334,
-      "loss": 8.0452,
-      "step": 8
-    },
-    {
-      "epoch": 1.6,
-      "grad_norm": 2.6754252910614014,
-      "learning_rate": 0.00011666666666666668,
-      "loss": 7.8375,
-      "step": 10
+      "loss": 9.1456,
+      "step": 4
     },
     {
-      "epoch": 1.92,
-      "grad_norm": 1.9012829065322876,
+      "epoch": 1.7777777777777777,
+      "grad_norm": 2.0616180896759033,
       "learning_rate": 0.0001,
-      "loss": 7.6614,
-      "step": 12
-    },
-    {
-      "epoch": 2.24,
-      "grad_norm": 1.1993588209152222,
-      "learning_rate": 8.333333333333334e-05,
-      "loss": 7.4492,
-      "step": 14
+      "loss": 8.731,
+      "step": 6
     },
     {
-      "epoch": 2.56,
-      "grad_norm": 1.133289098739624,
+      "epoch": 2.3703703703703702,
+      "grad_norm": 2.349423885345459,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 7.3398,
-      "step": 16
-    },
-    {
-      "epoch": 2.88,
-      "grad_norm": 1.9940211772918701,
-      "learning_rate": 5e-05,
-      "loss": 7.1652,
-      "step": 18
+      "loss": 8.2574,
+      "step": 8
     },
     {
-      "epoch": 3.2,
-      "grad_norm": 1.1616061925888062,
+      "epoch": 2.962962962962963,
+      "grad_norm": 1.723225474357605,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 7.1986,
-      "step": 20
-    },
-    {
-      "epoch": 3.52,
-      "grad_norm": 1.421707272529602,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 7.1556,
-      "step": 22
+      "loss": 7.9055,
+      "step": 10
     },
     {
-      "epoch": 3.84,
-      "grad_norm": 0.9798431396484375,
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.465832233428955,
       "learning_rate": 0.0,
-      "loss": 7.0969,
-      "step": 24
+      "loss": 7.7976,
+      "step": 12
     },
     {
-      "epoch": 3.84,
-      "step": 24,
-      "total_flos": 124389176652576.0,
-      "train_loss": 7.826913595199585,
-      "train_runtime": 120.6463,
-      "train_samples_per_second": 3.315,
-      "train_steps_per_second": 0.199
+      "epoch": 3.5555555555555554,
+      "step": 12,
+      "total_flos": 69844190099040.0,
+      "train_loss": 8.555556774139404,
+      "train_runtime": 79.8982,
+      "train_samples_per_second": 2.703,
+      "train_steps_per_second": 0.15
     }
   ],
   "logging_steps": 2,
-  "max_steps": 24,
+  "max_steps": 12,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
@@ -119,7 +77,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 124389176652576.0,
+  "total_flos": 69844190099040.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null