End of training

README.md

@@ -2,6 +2,8 @@
 license: cc-by-nc-4.0
 library_name: peft
 tags:
+- text-to-audio
+- ylacombe/tiny-punk
 - generated_from_trainer
 base_model: facebook/musicgen-melody
 model-index:
@@ -14,7 +16,10 @@ 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 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 - DEFAULT dataset.
+It achieves the following results on the evaluation set:
+- Loss: 5.0972
+- Clap: 0.0238
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 3.0,
-    "eval_clap": -0.018694762140512466,
-    "eval_loss": 4.857634544372559,
-    "eval_runtime": 150.0944,
-    "eval_samples": 8,
-    "eval_samples_per_second": 0.053,
-    "eval_steps_per_second": 0.053,
-    "total_flos": 415739226780.0,
-    "train_loss": 2.6683399379253387,
-    "train_runtime": 15.8841,
-    "train_samples": 5,
-    "train_samples_per_second": 1.259,
-    "train_steps_per_second": 0.252
+    "epoch": 4.0,
+    "eval_clap": 0.023790106177330017,
+    "eval_loss": 5.097151279449463,
+    "eval_runtime": 30.31,
+    "eval_samples": 5,
+    "eval_samples_per_second": 0.165,
+    "eval_steps_per_second": 0.033,
+    "total_flos": 110863793808.0,
+    "train_loss": 1.1294779181480408,
+    "train_runtime": 21.3494,
+    "train_samples": 1,
+    "train_samples_per_second": 0.187,
+    "train_steps_per_second": 0.187
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.0,
-    "eval_clap": -0.018694762140512466,
-    "eval_loss": 4.857634544372559,
-    "eval_runtime": 150.0944,
-    "eval_samples": 8,
-    "eval_samples_per_second": 0.053,
-    "eval_steps_per_second": 0.053
+    "epoch": 4.0,
+    "eval_clap": 0.023790106177330017,
+    "eval_loss": 5.097151279449463,
+    "eval_runtime": 30.31,
+    "eval_samples": 5,
+    "eval_samples_per_second": 0.165,
+    "eval_steps_per_second": 0.033
 }

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.0,
-    "total_flos": 415739226780.0,
-    "train_loss": 2.6683399379253387,
-    "train_runtime": 15.8841,
-    "train_samples": 5,
-    "train_samples_per_second": 1.259,
-    "train_steps_per_second": 0.252
+    "epoch": 4.0,
+    "total_flos": 110863793808.0,
+    "train_loss": 1.1294779181480408,
+    "train_runtime": 21.3494,
+    "train_samples": 1,
+    "train_samples_per_second": 0.187,
+    "train_steps_per_second": 0.187
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
+  "epoch": 4.0,
   "eval_steps": 25,
   "global_step": 4,
   "is_hyper_param_search": false,
@@ -10,40 +10,40 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.6173375248908997,
+      "grad_norm": 30085.0234375,
       "learning_rate": 0.00015000000000000001,
-      "loss": 3.628,
+      "loss": 1.1752,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.6236879825592041,
+      "grad_norm": 28943.8359375,
       "learning_rate": 0.0001,
-      "loss": 3.5988,
+      "loss": 1.1398,
       "step": 2
     },
     {
-      "epoch": 2.6666666666666665,
-      "grad_norm": 0.47706061601638794,
+      "epoch": 3.0,
+      "grad_norm": 29536.12890625,
       "learning_rate": 5e-05,
-      "loss": 2.3147,
+      "loss": 1.102,
       "step": 3
     },
     {
-      "epoch": 3.0,
-      "grad_norm": 0.338273286819458,
+      "epoch": 4.0,
+      "grad_norm": 34210.921875,
       "learning_rate": 0.0,
-      "loss": 1.1319,
+      "loss": 1.1008,
       "step": 4
     },
     {
-      "epoch": 3.0,
+      "epoch": 4.0,
       "step": 4,
-      "total_flos": 415739226780.0,
-      "train_loss": 2.6683399379253387,
-      "train_runtime": 15.8841,
-      "train_samples_per_second": 1.259,
-      "train_steps_per_second": 0.252
+      "total_flos": 110863793808.0,
+      "train_loss": 1.1294779181480408,
+      "train_runtime": 21.3494,
+      "train_samples_per_second": 0.187,
+      "train_steps_per_second": 0.187
     }
   ],
   "logging_steps": 1.0,
@@ -57,14 +57,14 @@
         "should_epoch_stop": false,
         "should_evaluate": false,
         "should_log": false,
-        "should_save": true,
-        "should_training_stop": true
+        "should_save": false,
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 415739226780.0,
-  "train_batch_size": 2,
+  "total_flos": 110863793808.0,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }