End of training

README.md

@@ -1,5 +1,8 @@
 ---
+base_model: ylacombe/musicgen-melody
 tags:
+- text-to-audio
+- patrickvonplaten/bella_ciao
 - generated_from_trainer
 datasets:
 - audiofolder
@@ -13,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # musicgen-melody-bella-ciao
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the audiofolder dataset.
+This model is a fine-tuned version of [ylacombe/musicgen-melody](https://huggingface.co/ylacombe/musicgen-melody) on the PATRICKVONPLATEN/BELLA_CIAO - DEFAULT dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 7.3888854471842444,
-    "train_runtime": 149.4092,
-    "train_samples": 30,
-    "train_samples_per_second": 2.008,
-    "train_steps_per_second": 2.008
+    "train_loss": 8.267948303222656,
+    "train_runtime": 72.315,
+    "train_samples": 10,
+    "train_samples_per_second": 1.383,
+    "train_steps_per_second": 1.383
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 7.3888854471842444,
-    "train_runtime": 149.4092,
-    "train_samples": 30,
-    "train_samples_per_second": 2.008,
-    "train_steps_per_second": 2.008
+    "train_loss": 8.267948303222656,
+    "train_runtime": 72.315,
+    "train_samples": 10,
+    "train_samples_per_second": 1.383,
+    "train_steps_per_second": 1.383
 }

trainer_state.json

@@ -3,95 +3,47 @@
   "best_model_checkpoint": null,
   "epoch": 10.0,
   "eval_steps": 1000,
-  "global_step": 300,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 0.83,
-      "learning_rate": 9.2e-07,
-      "loss": 8.7857,
-      "step": 25
-    },
-    {
-      "epoch": 1.67,
-      "learning_rate": 8.366666666666667e-07,
-      "loss": 7.6115,
-      "step": 50
-    },
     {
       "epoch": 2.5,
-      "learning_rate": 7.533333333333332e-07,
-      "loss": 7.3706,
-      "step": 75
-    },
-    {
-      "epoch": 3.33,
-      "learning_rate": 6.733333333333333e-07,
-      "loss": 7.2883,
-      "step": 100
-    },
-    {
-      "epoch": 4.17,
-      "learning_rate": 5.9e-07,
-      "loss": 7.2673,
-      "step": 125
+      "learning_rate": 3.7999999999999996e-07,
+      "loss": 9.1576,
+      "step": 25
     },
     {
       "epoch": 5.0,
-      "learning_rate": 5.066666666666667e-07,
-      "loss": 7.2174,
-      "step": 150
-    },
-    {
-      "epoch": 5.83,
-      "learning_rate": 4.2333333333333334e-07,
-      "loss": 7.2089,
-      "step": 175
-    },
-    {
-      "epoch": 6.67,
-      "learning_rate": 3.4000000000000003e-07,
-      "loss": 7.1989,
-      "step": 200
+      "learning_rate": 2.55e-07,
+      "loss": 8.3016,
+      "step": 50
     },
     {
       "epoch": 7.5,
-      "learning_rate": 2.5666666666666666e-07,
-      "loss": 7.1801,
-      "step": 225
-    },
-    {
-      "epoch": 8.33,
-      "learning_rate": 1.7333333333333332e-07,
-      "loss": 7.1842,
-      "step": 250
-    },
-    {
-      "epoch": 9.17,
-      "learning_rate": 9e-08,
-      "loss": 7.1852,
-      "step": 275
+      "learning_rate": 1.3e-07,
+      "loss": 7.9075,
+      "step": 75
     },
     {
       "epoch": 10.0,
-      "learning_rate": 1e-08,
-      "loss": 7.1686,
-      "step": 300
+      "learning_rate": 5e-09,
+      "loss": 7.7051,
+      "step": 100
     },
     {
       "epoch": 10.0,
-      "step": 300,
+      "step": 100,
       "total_flos": 0.0,
-      "train_loss": 7.3888854471842444,
-      "train_runtime": 149.4092,
-      "train_samples_per_second": 2.008,
-      "train_steps_per_second": 2.008
+      "train_loss": 8.267948303222656,
+      "train_runtime": 72.315,
+      "train_samples_per_second": 1.383,
+      "train_steps_per_second": 1.383
     }
   ],
   "logging_steps": 25,
-  "max_steps": 300,
+  "max_steps": 100,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,