End of training

README.md

@@ -3,6 +3,8 @@ base_model: facebook/musicgen-large
 library_name: peft
 license: cc-by-nc-4.0
 tags:
+- text-to-audio
+- acid_data
 - generated_from_trainer
 model-index:
 - name: musicgen-large-lora-acid_techno_4ep_35da-colab
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # musicgen-large-lora-acid_techno_4ep_35da-colab
 
-This model is a fine-tuned version of [facebook/musicgen-large](https://huggingface.co/facebook/musicgen-large) on an unknown dataset.
+This model is a fine-tuned version of [facebook/musicgen-large](https://huggingface.co/facebook/musicgen-large) on the ShaySha/acid_data dataset.
 
 ## Model description
 

trainer_state.json

@@ -1,53 +1,81 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.6153846153846154,
+  "epoch": 2.888888888888889,
   "eval_steps": 500,
-  "global_step": 4,
+  "global_step": 8,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.6153846153846154,
-      "grad_norm": 1.1644846200942993,
-      "learning_rate": 0.00015000000000000001,
-      "loss": 8.9729,
+      "epoch": 0.4444444444444444,
+      "grad_norm": 1.1740930080413818,
+      "learning_rate": 0.000175,
+      "loss": 8.9875,
       "step": 1
     },
     {
-      "epoch": 1.6153846153846154,
-      "grad_norm": 1.4217939376831055,
-      "learning_rate": 0.0001,
-      "loss": 10.0559,
+      "epoch": 0.8888888888888888,
+      "grad_norm": 1.2762857675552368,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 8.9379,
       "step": 2
     },
     {
-      "epoch": 2.6153846153846154,
-      "grad_norm": 2.3738677501678467,
-      "learning_rate": 5e-05,
-      "loss": 10.0118,
+      "epoch": 1.0,
+      "grad_norm": 0.6354243159294128,
+      "learning_rate": 0.000125,
+      "loss": 2.2202,
       "step": 3
     },
     {
-      "epoch": 3.6153846153846154,
-      "grad_norm": 2.8046391010284424,
-      "learning_rate": 0.0,
-      "loss": 9.9665,
+      "epoch": 1.4444444444444444,
+      "grad_norm": 2.7923552989959717,
+      "learning_rate": 0.0001,
+      "loss": 8.8081,
       "step": 4
     },
     {
-      "epoch": 3.6153846153846154,
-      "step": 4,
-      "total_flos": 11516965239360.0,
-      "train_loss": 9.751765489578247,
-      "train_runtime": 35.6843,
-      "train_samples_per_second": 2.802,
-      "train_steps_per_second": 0.112
+      "epoch": 1.8888888888888888,
+      "grad_norm": 3.2708566188812256,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 8.7649,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0620002746582031,
+      "learning_rate": 5e-05,
+      "loss": 2.1849,
+      "step": 6
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 3.5511021614074707,
+      "learning_rate": 2.5e-05,
+      "loss": 8.6598,
+      "step": 7
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 3.999887466430664,
+      "learning_rate": 0.0,
+      "loss": 8.615,
+      "step": 8
+    },
+    {
+      "epoch": 2.888888888888889,
+      "step": 8,
+      "total_flos": 16781863634496.0,
+      "train_loss": 7.147286474704742,
+      "train_runtime": 50.3802,
+      "train_samples_per_second": 2.779,
+      "train_steps_per_second": 0.159
     }
   ],
   "logging_steps": 1,
-  "max_steps": 4,
+  "max_steps": 8,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
@@ -63,7 +91,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 11516965239360.0,
+  "total_flos": 16781863634496.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null