Training in progress, step 28

adapter_config.json

@@ -4,7 +4,7 @@
     "base_model_class": "MusicgenMelodyForConditionalGeneration",
     "parent_library": "transformers.models.musicgen_melody.modeling_musicgen_melody"
   },
-  "base_model_name_or_path": "facebook/musicgen-melody",
+  "base_model_name_or_path": "/data/data3/hangzeli/work_lab/workspace/multimodal/huggingface_model/musicgen-melody",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -13,32 +13,32 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 8,
+  "lora_alpha": 16,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "fc2",
-    "lm_heads.3",
-    "lm_heads.2",
-    "q_proj",
-    "embed_tokens.3",
     "lm_heads.0",
+    "v_proj",
+    "enc_to_dec_proj",
+    "embed_tokens.1",
     "audio_enc_to_dec_proj",
     "fc1",
-    "enc_to_dec_proj",
-    "embed_tokens.2",
+    "fc2",
+    "q_proj",
     "k_proj",
-    "embed_tokens.0",
-    "out_proj",
     "lm_heads.1",
-    "embed_tokens.1",
-    "v_proj"
+    "lm_heads.2",
+    "lm_heads.3",
+    "out_proj",
+    "embed_tokens.2",
+    "embed_tokens.3",
+    "embed_tokens.0"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9d3d4aeaad756a5f6fe85132368a98a076d7c00abd1c5b2211ccc997f6aa91d
-size 43594480
+oid sha256:edbe6968ce91e71f4bba96413aaec339c38f9d2726f1b5bdcda8c0c8a61744d9
+size 87103456

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.6721311475409837,
-    "train_loss": 7.743570498057774,
-    "train_runtime": 121.4704,
+    "train_loss": 7.744901963642666,
+    "train_runtime": 121.5966,
     "train_samples": 122,
-    "train_samples_per_second": 4.017,
-    "train_steps_per_second": 0.231
+    "train_samples_per_second": 4.013,
+    "train_steps_per_second": 0.23
 }

tokenizer.json

@@ -1,7 +1,14 @@
 {
   "version": "1.0",
   "truncation": null,
-  "padding": null,
+  "padding": {
+    "strategy": "BatchLongest",
+    "direction": "Right",
+    "pad_to_multiple_of": null,
+    "pad_id": 0,
+    "pad_type_id": 0,
+    "pad_token": "<pad>"
+  },
   "added_tokens": [
     {
       "id": 0,

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.6721311475409837,
-    "train_loss": 7.743570498057774,
-    "train_runtime": 121.4704,
+    "train_loss": 7.744901963642666,
+    "train_runtime": 121.5966,
     "train_samples": 122,
-    "train_samples_per_second": 4.017,
-    "train_steps_per_second": 0.231
+    "train_samples_per_second": 4.013,
+    "train_steps_per_second": 0.23
 }

trainer_state.json

@@ -10,110 +10,110 @@
   "log_history": [
     {
       "epoch": 0.26229508196721313,
-      "grad_norm": 1.0287491083145142,
+      "grad_norm": 0.9730982184410095,
       "learning_rate": 0.00018571428571428572,
-      "loss": 9.4306,
+      "loss": 9.4352,
       "step": 2
     },
     {
       "epoch": 0.5245901639344263,
-      "grad_norm": 1.8649983406066895,
+      "grad_norm": 1.8006277084350586,
       "learning_rate": 0.00017142857142857143,
-      "loss": 9.1591,
+      "loss": 9.1689,
       "step": 4
     },
     {
       "epoch": 0.7868852459016393,
-      "grad_norm": 1.8567523956298828,
+      "grad_norm": 1.831032395362854,
       "learning_rate": 0.00015714285714285716,
-      "loss": 8.7072,
+      "loss": 8.7485,
       "step": 6
     },
     {
       "epoch": 1.0491803278688525,
-      "grad_norm": 2.180053234100342,
+      "grad_norm": 2.2559938430786133,
       "learning_rate": 0.00014285714285714287,
-      "loss": 8.2864,
+      "loss": 8.3076,
       "step": 8
     },
     {
       "epoch": 1.3114754098360657,
-      "grad_norm": 2.318220853805542,
+      "grad_norm": 2.454803943634033,
       "learning_rate": 0.00012857142857142858,
-      "loss": 7.8084,
+      "loss": 7.8092,
       "step": 10
     },
     {
       "epoch": 1.5737704918032787,
-      "grad_norm": 2.61151385307312,
+      "grad_norm": 2.537230968475342,
       "learning_rate": 0.00011428571428571428,
-      "loss": 7.8655,
+      "loss": 7.8537,
       "step": 12
     },
     {
       "epoch": 1.8360655737704918,
-      "grad_norm": 2.6175003051757812,
+      "grad_norm": 2.46220326423645,
       "learning_rate": 0.0001,
-      "loss": 7.6385,
+      "loss": 7.5995,
       "step": 14
     },
     {
       "epoch": 2.098360655737705,
-      "grad_norm": 1.4860339164733887,
+      "grad_norm": 1.449030876159668,
       "learning_rate": 8.571428571428571e-05,
-      "loss": 7.0809,
+      "loss": 7.0811,
       "step": 16
     },
     {
       "epoch": 2.360655737704918,
-      "grad_norm": 1.5985082387924194,
+      "grad_norm": 1.5821515321731567,
       "learning_rate": 7.142857142857143e-05,
-      "loss": 7.2482,
+      "loss": 7.2538,
       "step": 18
     },
     {
       "epoch": 2.6229508196721314,
-      "grad_norm": 1.956580638885498,
+      "grad_norm": 1.8766202926635742,
       "learning_rate": 5.714285714285714e-05,
-      "loss": 7.0691,
+      "loss": 7.068,
       "step": 20
     },
     {
       "epoch": 2.8852459016393444,
-      "grad_norm": 1.449498176574707,
+      "grad_norm": 1.4384005069732666,
       "learning_rate": 4.2857142857142856e-05,
-      "loss": 7.0927,
+      "loss": 7.0955,
       "step": 22
     },
     {
       "epoch": 3.1475409836065573,
-      "grad_norm": 1.9451191425323486,
+      "grad_norm": 1.8369501829147339,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 6.8079,
+      "loss": 6.7993,
       "step": 24
     },
     {
       "epoch": 3.4098360655737707,
-      "grad_norm": 1.4124294519424438,
+      "grad_norm": 1.4276976585388184,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 7.0556,
+      "loss": 7.053,
       "step": 26
     },
     {
       "epoch": 3.6721311475409837,
-      "grad_norm": 1.6439225673675537,
+      "grad_norm": 1.627907156944275,
       "learning_rate": 0.0,
-      "loss": 7.1598,
+      "loss": 7.1552,
       "step": 28
     },
     {
       "epoch": 3.6721311475409837,
       "step": 28,
-      "total_flos": 131289516862176.0,
-      "train_loss": 7.743570498057774,
-      "train_runtime": 121.4704,
-      "train_samples_per_second": 4.017,
-      "train_steps_per_second": 0.231
+      "total_flos": 131381250733656.0,
+      "train_loss": 7.744901963642666,
+      "train_runtime": 121.5966,
+      "train_samples_per_second": 4.013,
+      "train_steps_per_second": 0.23
     }
   ],
   "logging_steps": 2,
@@ -133,7 +133,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 131289516862176.0,
+  "total_flos": 131381250733656.0,
   "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:a1637a4130ec7c578e47549d30e4fc16da4b978ca4e0d66b5de6e12e322f6ff7
+oid sha256:ecdfb3bc0586502d5167e8f227e1e825f93db0647c910a4157114e83c2f11d51
 size 5304