End of training

README.md

@@ -3,6 +3,8 @@ 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
@@ -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: 4.7421
+- Clap: -0.0067
 
 ## Model description
 

all_results.json

@@ -1,9 +1,15 @@
 {
-    "epoch": 3.764705882352941,
-    "total_flos": 3464493556500.0,
-    "train_loss": 8.955702781677246,
-    "train_runtime": 51.5041,
+    "epoch": 3.5555555555555554,
+    "eval_clap": -0.006718707736581564,
+    "eval_loss": 4.742068767547607,
+    "eval_runtime": 163.4795,
+    "eval_samples": 5,
+    "eval_samples_per_second": 0.031,
+    "eval_steps_per_second": 0.018,
+    "total_flos": 3298197865788.0,
+    "train_loss": 9.308886766433716,
+    "train_runtime": 46.4172,
     "train_samples": 33,
-    "train_samples_per_second": 2.563,
-    "train_steps_per_second": 0.155
+    "train_samples_per_second": 2.844,
+    "train_steps_per_second": 0.086
 }

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.5555555555555554,
+    "eval_clap": -0.006718707736581564,
+    "eval_loss": 4.742068767547607,
+    "eval_runtime": 163.4795,
+    "eval_samples": 5,
+    "eval_samples_per_second": 0.031,
+    "eval_steps_per_second": 0.018
+}

runs/Nov10_19-01-02_bdaa6bed38f5/events.out.tfevents.1731265565.bdaa6bed38f5.578166.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1cd8b2c304d3fc5cfe7cee6f51bc97080657bbf6824fa51393fe30a224d28d7
+size 401

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.764705882352941,
-    "total_flos": 3464493556500.0,
-    "train_loss": 8.955702781677246,
-    "train_runtime": 51.5041,
+    "epoch": 3.5555555555555554,
+    "total_flos": 3298197865788.0,
+    "train_loss": 9.308886766433716,
+    "train_runtime": 46.4172,
     "train_samples": 33,
-    "train_samples_per_second": 2.563,
-    "train_steps_per_second": 0.155
+    "train_samples_per_second": 2.844,
+    "train_steps_per_second": 0.086
 }

trainer_state.json

@@ -1,81 +1,53 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.764705882352941,
+  "epoch": 3.5555555555555554,
   "eval_steps": 25,
-  "global_step": 8,
+  "global_step": 4,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.47058823529411764,
-      "grad_norm": 9.738798141479492,
-      "learning_rate": 0.000175,
-      "loss": 9.5576,
-      "step": 1
-    },
-    {
-      "epoch": 0.9411764705882353,
-      "grad_norm": 9.771041870117188,
+      "epoch": 0.8888888888888888,
+      "grad_norm": 9.884089469909668,
       "learning_rate": 0.00015000000000000001,
-      "loss": 9.4171,
-      "step": 2
-    },
-    {
-      "epoch": 1.4117647058823528,
-      "grad_norm": 9.270547866821289,
-      "learning_rate": 0.000125,
-      "loss": 9.2248,
-      "step": 3
+      "loss": 9.5546,
+      "step": 1
     },
     {
-      "epoch": 1.8823529411764706,
-      "grad_norm": 10.51770305633545,
+      "epoch": 1.7777777777777777,
+      "grad_norm": 8.386604309082031,
       "learning_rate": 0.0001,
-      "loss": 8.9946,
-      "step": 4
-    },
-    {
-      "epoch": 2.3529411764705883,
-      "grad_norm": 13.01863956451416,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 8.8172,
-      "step": 5
+      "loss": 9.3703,
+      "step": 2
     },
     {
-      "epoch": 2.8235294117647056,
-      "grad_norm": 14.001367568969727,
+      "epoch": 2.6666666666666665,
+      "grad_norm": 9.898933410644531,
       "learning_rate": 5e-05,
-      "loss": 8.702,
-      "step": 6
-    },
-    {
-      "epoch": 3.2941176470588234,
-      "grad_norm": 17.041738510131836,
-      "learning_rate": 2.5e-05,
-      "loss": 8.5038,
-      "step": 7
+      "loss": 9.2303,
+      "step": 3
     },
     {
-      "epoch": 3.764705882352941,
-      "grad_norm": 17.74396514892578,
+      "epoch": 3.5555555555555554,
+      "grad_norm": 10.148322105407715,
       "learning_rate": 0.0,
-      "loss": 8.4284,
-      "step": 8
+      "loss": 9.0802,
+      "step": 4
     },
     {
-      "epoch": 3.764705882352941,
-      "step": 8,
-      "total_flos": 3464493556500.0,
-      "train_loss": 8.955702781677246,
-      "train_runtime": 51.5041,
-      "train_samples_per_second": 2.563,
-      "train_steps_per_second": 0.155
+      "epoch": 3.5555555555555554,
+      "step": 4,
+      "total_flos": 3298197865788.0,
+      "train_loss": 9.308886766433716,
+      "train_runtime": 46.4172,
+      "train_samples_per_second": 2.844,
+      "train_steps_per_second": 0.086
     }
   ],
   "logging_steps": 1.0,
-  "max_steps": 8,
+  "max_steps": 4,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
@@ -91,8 +63,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 3464493556500.0,
-  "train_batch_size": 2,
+  "total_flos": 3298197865788.0,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }