End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.9403973509933774,
-    "eval_accuracy": 0.8683473389355743,
-    "eval_loss": 0.4009244740009308,
-    "eval_runtime": 16.7576,
-    "eval_samples_per_second": 63.911,
-    "eval_steps_per_second": 1.014,
-    "total_flos": 2.897918409348219e+18,
-    "train_loss": 0.8824126113642443,
-    "train_runtime": 1162.7997,
-    "train_samples_per_second": 24.868,
-    "train_steps_per_second": 0.095
+    "epoch": 1.0,
+    "eval_accuracy": 0.8634920634920635,
+    "eval_loss": 0.44639232754707336,
+    "eval_runtime": 28.206,
+    "eval_samples_per_second": 67.007,
+    "eval_steps_per_second": 1.064,
+    "total_flos": 1.0941516662243328e+18,
+    "train_loss": 1.2471325397491455,
+    "train_runtime": 397.8778,
+    "train_samples_per_second": 26.918,
+    "train_steps_per_second": 0.106
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.9403973509933774,
-    "eval_accuracy": 0.8683473389355743,
-    "eval_loss": 0.4009244740009308,
-    "eval_runtime": 16.7576,
-    "eval_samples_per_second": 63.911,
-    "eval_steps_per_second": 1.014
+    "epoch": 1.0,
+    "eval_accuracy": 0.8634920634920635,
+    "eval_loss": 0.44639232754707336,
+    "eval_runtime": 28.206,
+    "eval_samples_per_second": 67.007,
+    "eval_steps_per_second": 1.064
 }

runs/Sep06_22-24-21_2de8c0b5efc2/events.out.tfevents.1725661898.2de8c0b5efc2.36.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8983b84d875ab0888b68e5a5f73192c6c50d04509d3951ddac0472a9a61f203
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.9403973509933774,
-    "total_flos": 2.897918409348219e+18,
-    "train_loss": 0.8824126113642443,
-    "train_runtime": 1162.7997,
-    "train_samples_per_second": 24.868,
-    "train_steps_per_second": 0.095
+    "epoch": 1.0,
+    "total_flos": 1.0941516662243328e+18,
+    "train_loss": 1.2471325397491455,
+    "train_runtime": 397.8778,
+    "train_samples_per_second": 26.918,
+    "train_steps_per_second": 0.106
 }

trainer_state.json

@@ -1,131 +1,64 @@
 {
-  "best_metric": 0.8683473389355743,
-  "best_model_checkpoint": "/kaggle/working/dinov2-base-finetuned-har/checkpoint-111",
-  "epoch": 2.9403973509933774,
+  "best_metric": 0.8634920634920635,
+  "best_model_checkpoint": "dinov2-base-finetuned-har/checkpoint-42",
+  "epoch": 1.0,
   "eval_steps": 500,
-  "global_step": 111,
+  "global_step": 42,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.26490066225165565,
-      "grad_norm": 80.4668197631836,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 2.2896,
+      "epoch": 0.23809523809523808,
+      "grad_norm": 44.80964660644531,
+      "learning_rate": 4.324324324324325e-05,
+      "loss": 2.2183,
       "step": 10
     },
     {
-      "epoch": 0.5298013245033113,
-      "grad_norm": 28.893081665039062,
-      "learning_rate": 4.595959595959596e-05,
-      "loss": 1.1524,
+      "epoch": 0.47619047619047616,
+      "grad_norm": 20.761051177978516,
+      "learning_rate": 2.9729729729729733e-05,
+      "loss": 1.1574,
       "step": 20
     },
     {
-      "epoch": 0.7947019867549668,
-      "grad_norm": 21.720983505249023,
-      "learning_rate": 4.0909090909090915e-05,
-      "loss": 0.9308,
+      "epoch": 0.7142857142857143,
+      "grad_norm": 28.136566162109375,
+      "learning_rate": 1.6216216216216218e-05,
+      "loss": 0.8864,
       "step": 30
     },
     {
-      "epoch": 0.9801324503311258,
-      "eval_accuracy": 0.8300653594771242,
-      "eval_loss": 0.5691878795623779,
-      "eval_runtime": 22.1989,
-      "eval_samples_per_second": 48.246,
-      "eval_steps_per_second": 0.766,
-      "step": 37
-    },
-    {
-      "epoch": 1.0596026490066226,
-      "grad_norm": 22.94610023498535,
-      "learning_rate": 3.5858585858585855e-05,
-      "loss": 0.863,
+      "epoch": 0.9523809523809523,
+      "grad_norm": 15.380955696105957,
+      "learning_rate": 2.702702702702703e-06,
+      "loss": 0.8084,
       "step": 40
     },
     {
-      "epoch": 1.3245033112582782,
-      "grad_norm": 24.597158432006836,
-      "learning_rate": 3.080808080808081e-05,
-      "loss": 0.7744,
-      "step": 50
-    },
-    {
-      "epoch": 1.589403973509934,
-      "grad_norm": 16.806970596313477,
-      "learning_rate": 2.575757575757576e-05,
-      "loss": 0.7287,
-      "step": 60
-    },
-    {
-      "epoch": 1.8543046357615895,
-      "grad_norm": 16.49362564086914,
-      "learning_rate": 2.070707070707071e-05,
-      "loss": 0.7052,
-      "step": 70
-    },
-    {
-      "epoch": 1.9867549668874172,
-      "eval_accuracy": 0.8468720821661998,
-      "eval_loss": 0.4805677831172943,
-      "eval_runtime": 15.9267,
-      "eval_samples_per_second": 67.246,
-      "eval_steps_per_second": 1.067,
-      "step": 75
-    },
-    {
-      "epoch": 2.119205298013245,
-      "grad_norm": 22.60504722595215,
-      "learning_rate": 1.565656565656566e-05,
-      "loss": 0.6016,
-      "step": 80
-    },
-    {
-      "epoch": 2.384105960264901,
-      "grad_norm": 15.832114219665527,
-      "learning_rate": 1.0606060606060607e-05,
-      "loss": 0.5896,
-      "step": 90
-    },
-    {
-      "epoch": 2.6490066225165565,
-      "grad_norm": 12.651689529418945,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.5613,
-      "step": 100
-    },
-    {
-      "epoch": 2.9139072847682117,
-      "grad_norm": 17.284223556518555,
-      "learning_rate": 5.050505050505052e-07,
-      "loss": 0.5414,
-      "step": 110
-    },
-    {
-      "epoch": 2.9403973509933774,
-      "eval_accuracy": 0.8683473389355743,
-      "eval_loss": 0.4009244740009308,
-      "eval_runtime": 15.8918,
-      "eval_samples_per_second": 67.393,
-      "eval_steps_per_second": 1.07,
-      "step": 111
+      "epoch": 1.0,
+      "eval_accuracy": 0.8634920634920635,
+      "eval_loss": 0.44639232754707336,
+      "eval_runtime": 28.1795,
+      "eval_samples_per_second": 67.07,
+      "eval_steps_per_second": 1.065,
+      "step": 42
     },
     {
-      "epoch": 2.9403973509933774,
-      "step": 111,
-      "total_flos": 2.897918409348219e+18,
-      "train_loss": 0.8824126113642443,
-      "train_runtime": 1162.7997,
-      "train_samples_per_second": 24.868,
-      "train_steps_per_second": 0.095
+      "epoch": 1.0,
+      "step": 42,
+      "total_flos": 1.0941516662243328e+18,
+      "train_loss": 1.2471325397491455,
+      "train_runtime": 397.8778,
+      "train_samples_per_second": 26.918,
+      "train_steps_per_second": 0.106
     }
   ],
   "logging_steps": 10,
-  "max_steps": 111,
+  "max_steps": 42,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 1,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -139,7 +72,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.897918409348219e+18,
+  "total_flos": 1.0941516662243328e+18,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null