Training in progress, step 56

README.md

@@ -7,14 +7,14 @@ tags:
 - dnb
 - generated_from_trainer
 model-index:
-- name: musicgen-melody-lora-dnb-colab-gs-8-lr-1e-05
+- name: musicgen-melody-lora-dnb-colab-gs-3-lr-0.0001
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# musicgen-melody-lora-dnb-colab-gs-8-lr-1e-05
+# musicgen-melody-lora-dnb-colab-gs-3-lr-0.0001
 
 This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the istvanj/ncs-dnb dataset.
 
@@ -35,7 +35,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
+- learning_rate: 0.0001
 - train_batch_size: 2
 - eval_batch_size: 8
 - seed: 42
@@ -43,7 +43,7 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 16
 - optimizer: Use adamw_torch with betas=(0.9,0.99) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 8
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results

adapter_config.json

@@ -23,22 +23,22 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "lm_heads.2",
-    "lm_heads.3",
-    "lm_heads.1",
+    "embed_tokens.1",
+    "fc1",
+    "lm_heads.0",
+    "out_proj",
     "embed_tokens.0",
     "q_proj",
-    "embed_tokens.2",
-    "embed_tokens.3",
-    "out_proj",
-    "fc1",
     "k_proj",
-    "embed_tokens.1",
+    "v_proj",
+    "lm_heads.2",
     "audio_enc_to_dec_proj",
+    "lm_heads.3",
     "fc2",
     "enc_to_dec_proj",
-    "lm_heads.0"
+    "embed_tokens.2",
+    "embed_tokens.3",
+    "lm_heads.1"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4ed8c891f22c4c9a7974cf2ad6e1bb84abbf140fe84f1372c0bd25752e57452
+oid sha256:1b3f8b7b1bbcdc6dc3f718146d672eb904d5c9c37afb3e9e0cf606c6eddfcdef
 size 87103456

trainer_state.json

@@ -1,223 +1,97 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 7.0,
+  "epoch": 2.6666666666666665,
   "eval_steps": 500,
-  "global_step": 56,
+  "global_step": 21,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.26666666666666666,
-      "grad_norm": 7.9753594398498535,
-      "learning_rate": 9.642857142857144e-06,
-      "loss": 9.7008,
+      "grad_norm": 7.614686012268066,
+      "learning_rate": 9.047619047619048e-05,
+      "loss": 9.6721,
       "step": 2
     },
     {
       "epoch": 0.5333333333333333,
-      "grad_norm": 8.494783401489258,
-      "learning_rate": 9.285714285714288e-06,
-      "loss": 9.6852,
+      "grad_norm": 10.107391357421875,
+      "learning_rate": 8.095238095238096e-05,
+      "loss": 9.4969,
       "step": 4
     },
     {
       "epoch": 0.8,
-      "grad_norm": 8.483810424804688,
-      "learning_rate": 8.92857142857143e-06,
-      "loss": 9.6854,
+      "grad_norm": 11.758378028869629,
+      "learning_rate": 7.142857142857143e-05,
+      "loss": 9.3036,
       "step": 6
     },
     {
       "epoch": 1.0,
-      "grad_norm": 4.711551666259766,
-      "learning_rate": 8.571428571428571e-06,
-      "loss": 7.2231,
+      "grad_norm": 8.324995994567871,
+      "learning_rate": 6.19047619047619e-05,
+      "loss": 6.8191,
       "step": 8
     },
     {
       "epoch": 1.2666666666666666,
-      "grad_norm": 8.524971961975098,
-      "learning_rate": 8.214285714285714e-06,
-      "loss": 9.6463,
+      "grad_norm": 15.543440818786621,
+      "learning_rate": 5.2380952380952384e-05,
+      "loss": 8.8411,
       "step": 10
     },
     {
       "epoch": 1.5333333333333332,
-      "grad_norm": 8.587374687194824,
-      "learning_rate": 8.035714285714286e-06,
-      "loss": 9.627,
+      "grad_norm": 16.783220291137695,
+      "learning_rate": 4.2857142857142856e-05,
+      "loss": 8.5329,
       "step": 12
     },
     {
       "epoch": 1.8,
-      "grad_norm": 9.312500953674316,
-      "learning_rate": 7.67857142857143e-06,
-      "loss": 9.5752,
+      "grad_norm": 19.108684539794922,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 8.2567,
       "step": 14
     },
     {
       "epoch": 2.0,
-      "grad_norm": 4.863954067230225,
-      "learning_rate": 7.321428571428572e-06,
-      "loss": 7.2151,
+      "grad_norm": 8.789897918701172,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 6.1492,
       "step": 16
     },
     {
       "epoch": 2.2666666666666666,
-      "grad_norm": 8.715095520019531,
-      "learning_rate": 7.1428571428571436e-06,
-      "loss": 9.5969,
+      "grad_norm": 17.98725128173828,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 8.0423,
       "step": 18
     },
     {
       "epoch": 2.533333333333333,
-      "grad_norm": 9.27696704864502,
-      "learning_rate": 6.785714285714287e-06,
-      "loss": 9.548,
+      "grad_norm": 14.338889122009277,
+      "learning_rate": 4.7619047619047615e-06,
+      "loss": 7.9496,
       "step": 20
     },
     {
-      "epoch": 2.8,
-      "grad_norm": 9.659817695617676,
-      "learning_rate": 6.4285714285714295e-06,
-      "loss": 9.5663,
-      "step": 22
-    },
-    {
-      "epoch": 3.0,
-      "grad_norm": 5.58745813369751,
-      "learning_rate": 6.071428571428571e-06,
-      "loss": 7.1335,
-      "step": 24
-    },
-    {
-      "epoch": 3.2666666666666666,
-      "grad_norm": 10.51404094696045,
-      "learning_rate": 5.7142857142857145e-06,
-      "loss": 9.5266,
-      "step": 26
-    },
-    {
-      "epoch": 3.533333333333333,
-      "grad_norm": 10.135522842407227,
-      "learning_rate": 5.357142857142857e-06,
-      "loss": 9.5168,
-      "step": 28
-    },
-    {
-      "epoch": 3.8,
-      "grad_norm": 10.956225395202637,
-      "learning_rate": 5e-06,
-      "loss": 9.5157,
-      "step": 30
-    },
-    {
-      "epoch": 4.0,
-      "grad_norm": 5.393774032592773,
-      "learning_rate": 4.642857142857144e-06,
-      "loss": 7.0646,
-      "step": 32
-    },
-    {
-      "epoch": 4.266666666666667,
-      "grad_norm": 10.676840782165527,
-      "learning_rate": 4.2857142857142855e-06,
-      "loss": 9.4731,
-      "step": 34
-    },
-    {
-      "epoch": 4.533333333333333,
-      "grad_norm": 10.430346488952637,
-      "learning_rate": 3.928571428571429e-06,
-      "loss": 9.4459,
-      "step": 36
-    },
-    {
-      "epoch": 4.8,
-      "grad_norm": 10.816131591796875,
-      "learning_rate": 3.5714285714285718e-06,
-      "loss": 9.4447,
-      "step": 38
-    },
-    {
-      "epoch": 5.0,
-      "grad_norm": 6.065415859222412,
-      "learning_rate": 3.2142857142857147e-06,
-      "loss": 7.0511,
-      "step": 40
-    },
-    {
-      "epoch": 5.266666666666667,
-      "grad_norm": 13.455761909484863,
-      "learning_rate": 2.8571428571428573e-06,
-      "loss": 9.4171,
-      "step": 42
-    },
-    {
-      "epoch": 5.533333333333333,
-      "grad_norm": 11.359918594360352,
-      "learning_rate": 2.5e-06,
-      "loss": 9.4188,
-      "step": 44
-    },
-    {
-      "epoch": 5.8,
-      "grad_norm": 13.576141357421875,
-      "learning_rate": 2.1428571428571427e-06,
-      "loss": 9.3739,
-      "step": 46
-    },
-    {
-      "epoch": 6.0,
-      "grad_norm": 6.167062759399414,
-      "learning_rate": 1.7857142857142859e-06,
-      "loss": 7.0424,
-      "step": 48
-    },
-    {
-      "epoch": 6.266666666666667,
-      "grad_norm": 13.524323463439941,
-      "learning_rate": 1.4285714285714286e-06,
-      "loss": 9.3775,
-      "step": 50
-    },
-    {
-      "epoch": 6.533333333333333,
-      "grad_norm": 20.678451538085938,
-      "learning_rate": 1.0714285714285714e-06,
-      "loss": 9.3592,
-      "step": 52
-    },
-    {
-      "epoch": 6.8,
-      "grad_norm": 12.042393684387207,
-      "learning_rate": 7.142857142857143e-07,
-      "loss": 9.3678,
-      "step": 54
-    },
-    {
-      "epoch": 7.0,
-      "grad_norm": 6.3625993728637695,
-      "learning_rate": 3.5714285714285716e-07,
-      "loss": 7.0451,
-      "step": 56
-    },
-    {
-      "epoch": 7.0,
-      "step": 56,
-      "total_flos": 216424602812184.0,
-      "train_loss": 8.915825281824384,
-      "train_runtime": 1374.5108,
-      "train_samples_per_second": 0.698,
-      "train_steps_per_second": 0.041
+      "epoch": 2.6666666666666665,
+      "step": 21,
+      "total_flos": 82187025809664.0,
+      "train_loss": 8.28742908296131,
+      "train_runtime": 577.7205,
+      "train_samples_per_second": 0.623,
+      "train_steps_per_second": 0.036
     }
   ],
   "logging_steps": 2,
-  "max_steps": 56,
+  "max_steps": 21,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -231,7 +105,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 216424602812184.0,
+  "total_flos": 82187025809664.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:778dc524f55afd59323688704b0726b11953ccd3eb2b83634379c146f19e65d7
+oid sha256:0c34c13b6fa5c21e9835c2ac37db1309b556f58be71c1278a44d7a2d2fd254c4
 size 5560