End of training

README.md

@@ -3,6 +3,8 @@ base_model: facebook/musicgen-small
 library_name: peft
 license: cc-by-nc-4.0
 tags:
+- text-to-audio
+- ogbanugot/musicgen
 - generated_from_trainer
 model-index:
 - name: musicgen-small-lora-afrobeats
@@ -15,7 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](None)
 # musicgen-small-lora-afrobeats
 
-This model is a fine-tuned version of [facebook/musicgen-small](https://huggingface.co/facebook/musicgen-small) on an unknown dataset.
+This model is a fine-tuned version of [facebook/musicgen-small](https://huggingface.co/facebook/musicgen-small) on the OGBANUGOT/MUSICGEN - DEFAULT dataset.
+It achieves the following results on the evaluation set:
+- Loss: 5.1124
+- Clap: 0.0623
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 3.84,
-    "eval_clap": 0.04958033934235573,
-    "eval_loss": 5.347536087036133,
-    "eval_runtime": 133.8795,
+    "eval_clap": 0.06234043464064598,
+    "eval_loss": 5.112404823303223,
+    "eval_runtime": 134.802,
     "eval_samples": 8,
-    "eval_samples_per_second": 0.06,
-    "eval_steps_per_second": 0.06,
-    "total_flos": 22357590798924.0,
-    "train_loss": 8.493526140848795,
-    "train_runtime": 47.9357,
+    "eval_samples_per_second": 0.059,
+    "eval_steps_per_second": 0.059,
+    "total_flos": 24492388886460.0,
+    "train_loss": 8.339014927546183,
+    "train_runtime": 59.862,
     "train_samples": 49,
-    "train_samples_per_second": 4.089,
-    "train_steps_per_second": 0.25
+    "train_samples_per_second": 3.274,
+    "train_steps_per_second": 0.2
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 3.84,
-    "eval_clap": 0.04958033934235573,
-    "eval_loss": 5.347536087036133,
-    "eval_runtime": 133.8795,
+    "eval_clap": 0.06234043464064598,
+    "eval_loss": 5.112404823303223,
+    "eval_runtime": 134.802,
     "eval_samples": 8,
-    "eval_samples_per_second": 0.06,
-    "eval_steps_per_second": 0.06
+    "eval_samples_per_second": 0.059,
+    "eval_steps_per_second": 0.059
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 3.84,
-    "total_flos": 22357590798924.0,
-    "train_loss": 8.493526140848795,
-    "train_runtime": 47.9357,
+    "total_flos": 24492388886460.0,
+    "train_loss": 8.339014927546183,
+    "train_runtime": 59.862,
     "train_samples": 49,
-    "train_samples_per_second": 4.089,
-    "train_steps_per_second": 0.25
+    "train_samples_per_second": 3.274,
+    "train_steps_per_second": 0.2
 }

trainer_state.json

@@ -10,96 +10,96 @@
   "log_history": [
     {
       "epoch": 0.32,
-      "grad_norm": 1.3747080564498901,
+      "grad_norm": 2.4018948078155518,
       "learning_rate": 0.00018333333333333334,
-      "loss": 9.5405,
+      "loss": 9.796,
       "step": 1
     },
     {
       "epoch": 0.64,
-      "grad_norm": 1.4412273168563843,
+      "grad_norm": 2.362989902496338,
       "learning_rate": 0.0001666666666666667,
-      "loss": 9.4519,
+      "loss": 9.5629,
       "step": 2
     },
     {
       "epoch": 0.96,
-      "grad_norm": 1.961427092552185,
+      "grad_norm": 3.1423513889312744,
       "learning_rate": 0.00015000000000000001,
-      "loss": 9.1814,
+      "loss": 9.2115,
       "step": 3
     },
     {
       "epoch": 1.28,
-      "grad_norm": 2.2182157039642334,
+      "grad_norm": 2.852221965789795,
       "learning_rate": 0.00013333333333333334,
-      "loss": 8.8873,
+      "loss": 8.7974,
       "step": 4
     },
     {
       "epoch": 1.6,
-      "grad_norm": 2.371889591217041,
+      "grad_norm": 3.004337787628174,
       "learning_rate": 0.00011666666666666668,
-      "loss": 8.6062,
+      "loss": 8.4192,
       "step": 5
     },
     {
       "epoch": 1.92,
-      "grad_norm": 1.932769536972046,
+      "grad_norm": 2.4463579654693604,
       "learning_rate": 0.0001,
-      "loss": 8.3597,
+      "loss": 8.0889,
       "step": 6
     },
     {
       "epoch": 2.24,
-      "grad_norm": 1.71647047996521,
+      "grad_norm": 2.03631854057312,
       "learning_rate": 8.333333333333334e-05,
-      "loss": 8.2076,
+      "loss": 7.8931,
       "step": 7
     },
     {
       "epoch": 2.56,
-      "grad_norm": 1.4216786623001099,
+      "grad_norm": 1.7664697170257568,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 8.0357,
+      "loss": 7.7725,
       "step": 8
     },
     {
       "epoch": 2.88,
-      "grad_norm": 1.3276846408843994,
+      "grad_norm": 2.012162685394287,
       "learning_rate": 5e-05,
-      "loss": 7.9892,
+      "loss": 7.7224,
       "step": 9
     },
     {
       "epoch": 3.2,
-      "grad_norm": 1.295423150062561,
+      "grad_norm": 1.7454917430877686,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 7.9392,
+      "loss": 7.5964,
       "step": 10
     },
     {
       "epoch": 3.52,
-      "grad_norm": 1.1277278661727905,
+      "grad_norm": 1.905915379524231,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 7.8902,
+      "loss": 7.6713,
       "step": 11
     },
     {
       "epoch": 3.84,
-      "grad_norm": 1.0911381244659424,
+      "grad_norm": 1.5737876892089844,
       "learning_rate": 0.0,
-      "loss": 7.8333,
+      "loss": 7.5365,
       "step": 12
     },
     {
       "epoch": 3.84,
       "step": 12,
-      "total_flos": 22357590798924.0,
-      "train_loss": 8.493526140848795,
-      "train_runtime": 47.9357,
-      "train_samples_per_second": 4.089,
-      "train_steps_per_second": 0.25
+      "total_flos": 24492388886460.0,
+      "train_loss": 8.339014927546183,
+      "train_runtime": 59.862,
+      "train_samples_per_second": 3.274,
+      "train_steps_per_second": 0.2
     }
   ],
   "logging_steps": 1.0,
@@ -119,7 +119,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 22357590798924.0,
+  "total_flos": 24492388886460.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null