ThyroidTumorClassification

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 19.73,
-    "eval_f1": 0.9312977099236641,
-    "eval_loss": 0.5987359285354614,
-    "eval_runtime": 0.4648,
-    "eval_samples_per_second": 176.425,
-    "eval_steps_per_second": 6.455,
-    "total_flos": 1.620732679243776e+17,
-    "train_loss": 0.8578881025314331,
-    "train_runtime": 102.9255,
-    "train_samples_per_second": 63.347,
-    "train_steps_per_second": 0.389
+    "epoch": 30.0,
+    "eval_f1": 0.8837209302325582,
+    "eval_loss": 0.5682619214057922,
+    "eval_runtime": 0.6789,
+    "eval_samples_per_second": 139.94,
+    "eval_steps_per_second": 4.419,
+    "total_flos": 2.8570124903878656e+17,
+    "train_loss": 0.5013438595665826,
+    "train_runtime": 222.831,
+    "train_samples_per_second": 51.025,
+    "train_steps_per_second": 0.404
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 19.73,
-    "eval_f1": 0.9312977099236641,
-    "eval_loss": 0.5987359285354614,
-    "eval_runtime": 0.4648,
-    "eval_samples_per_second": 176.425,
-    "eval_steps_per_second": 6.455
+    "epoch": 30.0,
+    "eval_f1": 0.8837209302325582,
+    "eval_loss": 0.5682619214057922,
+    "eval_runtime": 0.6789,
+    "eval_samples_per_second": 139.94,
+    "eval_steps_per_second": 4.419
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ce7c3b86954dad0996ccf49be4284947b73504f1521f15cd9a889bf12083e90
+oid sha256:5f5768f543feec39b661a0d3347394478b69fe2fb6010928ea157000d3c03bb2
 size 111347349

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 19.73,
-    "total_flos": 1.620732679243776e+17,
-    "train_loss": 0.8578881025314331,
-    "train_runtime": 102.9255,
-    "train_samples_per_second": 63.347,
-    "train_steps_per_second": 0.389
+    "epoch": 30.0,
+    "total_flos": 2.8570124903878656e+17,
+    "train_loss": 0.5013438595665826,
+    "train_runtime": 222.831,
+    "train_samples_per_second": 51.025,
+    "train_steps_per_second": 0.404
 }

trainer_state.json

@@ -1,211 +1,313 @@
 {
-  "best_metric": 0.9312977099236641,
-  "best_model_checkpoint": "Cvt-finetuned-thyroid/checkpoint-36",
-  "epoch": 19.727272727272727,
-  "global_step": 40,
+  "best_metric": 0.8837209302325582,
+  "best_model_checkpoint": "Cvt-finetuned-thyroid/checkpoint-27",
+  "epoch": 30.0,
+  "global_step": 90,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.73,
-      "eval_f1": 0.08695652173913043,
-      "eval_loss": 0.729124128818512,
-      "eval_runtime": 0.5602,
-      "eval_samples_per_second": 146.387,
-      "eval_steps_per_second": 5.356,
-      "step": 2
-    },
-    {
-      "epoch": 1.73,
-      "eval_f1": 0.24,
-      "eval_loss": 0.715528666973114,
-      "eval_runtime": 0.4576,
-      "eval_samples_per_second": 179.212,
-      "eval_steps_per_second": 6.557,
-      "step": 4
-    },
-    {
-      "epoch": 2.73,
-      "eval_f1": 0.4651162790697675,
-      "eval_loss": 0.6962855458259583,
-      "eval_runtime": 0.4553,
-      "eval_samples_per_second": 180.089,
-      "eval_steps_per_second": 6.589,
-      "step": 6
+      "epoch": 1.0,
+      "eval_f1": 0.5052631578947367,
+      "eval_loss": 0.687919557094574,
+      "eval_runtime": 0.6589,
+      "eval_samples_per_second": 144.182,
+      "eval_steps_per_second": 4.553,
+      "step": 3
     },
     {
-      "epoch": 3.73,
-      "eval_f1": 0.6732673267326733,
-      "eval_loss": 0.6805053353309631,
-      "eval_runtime": 0.456,
-      "eval_samples_per_second": 179.84,
-      "eval_steps_per_second": 6.58,
-      "step": 8
+      "epoch": 2.0,
+      "eval_f1": 0.673076923076923,
+      "eval_loss": 0.6748365759849548,
+      "eval_runtime": 0.6341,
+      "eval_samples_per_second": 149.822,
+      "eval_steps_per_second": 4.731,
+      "step": 6
     },
     {
-      "epoch": 4.73,
-      "eval_f1": 0.8035714285714285,
-      "eval_loss": 0.6671938896179199,
-      "eval_runtime": 0.4582,
-      "eval_samples_per_second": 178.951,
-      "eval_steps_per_second": 6.547,
-      "step": 10
+      "epoch": 3.0,
+      "eval_f1": 0.8034188034188033,
+      "eval_loss": 0.6552020311355591,
+      "eval_runtime": 0.6415,
+      "eval_samples_per_second": 148.089,
+      "eval_steps_per_second": 4.676,
+      "step": 9
     },
     {
-      "epoch": 5.73,
-      "eval_f1": 0.8666666666666666,
-      "eval_loss": 0.6562464237213135,
-      "eval_runtime": 0.4523,
-      "eval_samples_per_second": 181.301,
-      "eval_steps_per_second": 6.633,
+      "epoch": 4.0,
+      "eval_f1": 0.88,
+      "eval_loss": 0.6357027888298035,
+      "eval_runtime": 0.6465,
+      "eval_samples_per_second": 146.936,
+      "eval_steps_per_second": 4.64,
       "step": 12
     },
     {
-      "epoch": 6.73,
-      "eval_f1": 0.8688524590163934,
-      "eval_loss": 0.6468915343284607,
-      "eval_runtime": 0.4563,
-      "eval_samples_per_second": 179.688,
-      "eval_steps_per_second": 6.574,
-      "step": 14
+      "epoch": 5.0,
+      "eval_f1": 0.8615384615384616,
+      "eval_loss": 0.6192407011985779,
+      "eval_runtime": 0.6387,
+      "eval_samples_per_second": 148.729,
+      "eval_steps_per_second": 4.697,
+      "step": 15
     },
     {
-      "epoch": 7.73,
-      "eval_f1": 0.9047619047619049,
-      "eval_loss": 0.6390055418014526,
-      "eval_runtime": 0.4554,
-      "eval_samples_per_second": 180.045,
-      "eval_steps_per_second": 6.587,
-      "step": 16
+      "epoch": 6.0,
+      "eval_f1": 0.8769230769230769,
+      "eval_loss": 0.604374349117279,
+      "eval_runtime": 0.645,
+      "eval_samples_per_second": 147.298,
+      "eval_steps_per_second": 4.652,
+      "step": 18
     },
     {
-      "epoch": 8.73,
-      "eval_f1": 0.9218749999999999,
-      "eval_loss": 0.6321854591369629,
-      "eval_runtime": 0.4533,
-      "eval_samples_per_second": 180.88,
-      "eval_steps_per_second": 6.618,
-      "step": 18
+      "epoch": 7.0,
+      "eval_f1": 0.8769230769230769,
+      "eval_loss": 0.5909456014633179,
+      "eval_runtime": 0.6526,
+      "eval_samples_per_second": 145.576,
+      "eval_steps_per_second": 4.597,
+      "step": 21
     },
     {
-      "epoch": 9.73,
-      "eval_f1": 0.9218749999999999,
-      "eval_loss": 0.6260910034179688,
-      "eval_runtime": 0.4585,
-      "eval_samples_per_second": 178.853,
-      "eval_steps_per_second": 6.543,
-      "step": 20
+      "epoch": 8.0,
+      "eval_f1": 0.8769230769230769,
+      "eval_loss": 0.5791701674461365,
+      "eval_runtime": 0.6491,
+      "eval_samples_per_second": 146.347,
+      "eval_steps_per_second": 4.621,
+      "step": 24
     },
     {
-      "epoch": 10.73,
-      "eval_f1": 0.9147286821705426,
-      "eval_loss": 0.6207503080368042,
-      "eval_runtime": 0.4613,
-      "eval_samples_per_second": 177.741,
-      "eval_steps_per_second": 6.503,
-      "step": 22
+      "epoch": 8.33,
+      "learning_rate": 8.024691358024692e-06,
+      "loss": 0.638,
+      "step": 25
     },
     {
-      "epoch": 11.73,
-      "eval_f1": 0.9147286821705426,
-      "eval_loss": 0.6160098910331726,
-      "eval_runtime": 0.4683,
-      "eval_samples_per_second": 175.09,
-      "eval_steps_per_second": 6.406,
-      "step": 24
+      "epoch": 9.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5682619214057922,
+      "eval_runtime": 0.6497,
+      "eval_samples_per_second": 146.222,
+      "eval_steps_per_second": 4.618,
+      "step": 27
     },
     {
-      "epoch": 12.36,
-      "learning_rate": 4.166666666666667e-06,
-      "loss": 0.8937,
-      "step": 25
+      "epoch": 10.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5576053261756897,
+      "eval_runtime": 0.653,
+      "eval_samples_per_second": 145.478,
+      "eval_steps_per_second": 4.594,
+      "step": 30
+    },
+    {
+      "epoch": 11.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5474461913108826,
+      "eval_runtime": 0.651,
+      "eval_samples_per_second": 145.928,
+      "eval_steps_per_second": 4.608,
+      "step": 33
     },
     {
-      "epoch": 12.73,
-      "eval_f1": 0.9147286821705426,
-      "eval_loss": 0.6118525862693787,
-      "eval_runtime": 0.4721,
-      "eval_samples_per_second": 173.678,
-      "eval_steps_per_second": 6.354,
-      "step": 26
+      "epoch": 12.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5376133322715759,
+      "eval_runtime": 0.6594,
+      "eval_samples_per_second": 144.075,
+      "eval_steps_per_second": 4.55,
+      "step": 36
     },
     {
-      "epoch": 13.73,
-      "eval_f1": 0.9147286821705426,
-      "eval_loss": 0.6081259846687317,
-      "eval_runtime": 0.4553,
-      "eval_samples_per_second": 180.118,
-      "eval_steps_per_second": 6.59,
-      "step": 28
+      "epoch": 13.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5282223224639893,
+      "eval_runtime": 0.6631,
+      "eval_samples_per_second": 143.267,
+      "eval_steps_per_second": 4.524,
+      "step": 39
     },
     {
-      "epoch": 14.73,
-      "eval_f1": 0.923076923076923,
-      "eval_loss": 0.6047901511192322,
-      "eval_runtime": 0.4555,
-      "eval_samples_per_second": 180.011,
-      "eval_steps_per_second": 6.586,
-      "step": 30
+      "epoch": 14.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5196569561958313,
+      "eval_runtime": 0.6551,
+      "eval_samples_per_second": 145.016,
+      "eval_steps_per_second": 4.579,
+      "step": 42
     },
     {
-      "epoch": 15.73,
-      "eval_f1": 0.923076923076923,
-      "eval_loss": 0.6022080183029175,
-      "eval_runtime": 0.4519,
-      "eval_samples_per_second": 181.441,
-      "eval_steps_per_second": 6.638,
-      "step": 32
+      "epoch": 15.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5118635296821594,
+      "eval_runtime": 0.6658,
+      "eval_samples_per_second": 142.69,
+      "eval_steps_per_second": 4.506,
+      "step": 45
     },
     {
-      "epoch": 16.73,
-      "eval_f1": 0.923076923076923,
-      "eval_loss": 0.6001817584037781,
-      "eval_runtime": 0.4544,
-      "eval_samples_per_second": 180.453,
-      "eval_steps_per_second": 6.602,
-      "step": 34
+      "epoch": 16.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.5053044557571411,
+      "eval_runtime": 0.6565,
+      "eval_samples_per_second": 144.71,
+      "eval_steps_per_second": 4.57,
+      "step": 48
     },
     {
-      "epoch": 17.73,
-      "eval_f1": 0.9312977099236641,
-      "eval_loss": 0.5987359285354614,
-      "eval_runtime": 0.4559,
-      "eval_samples_per_second": 179.858,
-      "eval_steps_per_second": 6.58,
-      "step": 36
+      "epoch": 16.67,
+      "learning_rate": 4.938271604938272e-06,
+      "loss": 0.4926,
+      "step": 50
+    },
+    {
+      "epoch": 17.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.49933797121047974,
+      "eval_runtime": 0.6612,
+      "eval_samples_per_second": 143.679,
+      "eval_steps_per_second": 4.537,
+      "step": 51
+    },
+    {
+      "epoch": 18.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4940946102142334,
+      "eval_runtime": 0.661,
+      "eval_samples_per_second": 143.726,
+      "eval_steps_per_second": 4.539,
+      "step": 54
+    },
+    {
+      "epoch": 19.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4895910918712616,
+      "eval_runtime": 0.6623,
+      "eval_samples_per_second": 143.44,
+      "eval_steps_per_second": 4.53,
+      "step": 57
+    },
+    {
+      "epoch": 20.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4853971004486084,
+      "eval_runtime": 0.6837,
+      "eval_samples_per_second": 138.946,
+      "eval_steps_per_second": 4.388,
+      "step": 60
+    },
+    {
+      "epoch": 21.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.48188260197639465,
+      "eval_runtime": 0.6669,
+      "eval_samples_per_second": 142.45,
+      "eval_steps_per_second": 4.498,
+      "step": 63
+    },
+    {
+      "epoch": 22.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4790306091308594,
+      "eval_runtime": 0.6675,
+      "eval_samples_per_second": 142.322,
+      "eval_steps_per_second": 4.494,
+      "step": 66
+    },
+    {
+      "epoch": 23.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.47673219442367554,
+      "eval_runtime": 0.6755,
+      "eval_samples_per_second": 140.639,
+      "eval_steps_per_second": 4.441,
+      "step": 69
+    },
+    {
+      "epoch": 24.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4748006761074066,
+      "eval_runtime": 0.6675,
+      "eval_samples_per_second": 142.313,
+      "eval_steps_per_second": 4.494,
+      "step": 72
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 1.8518518518518519e-06,
+      "loss": 0.4289,
+      "step": 75
+    },
+    {
+      "epoch": 25.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4731859564781189,
+      "eval_runtime": 0.6627,
+      "eval_samples_per_second": 143.353,
+      "eval_steps_per_second": 4.527,
+      "step": 75
+    },
+    {
+      "epoch": 26.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4718676507472992,
+      "eval_runtime": 0.6719,
+      "eval_samples_per_second": 141.397,
+      "eval_steps_per_second": 4.465,
+      "step": 78
+    },
+    {
+      "epoch": 27.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4708217680454254,
+      "eval_runtime": 0.68,
+      "eval_samples_per_second": 139.706,
+      "eval_steps_per_second": 4.412,
+      "step": 81
+    },
+    {
+      "epoch": 28.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.47011542320251465,
+      "eval_runtime": 0.6739,
+      "eval_samples_per_second": 140.966,
+      "eval_steps_per_second": 4.452,
+      "step": 84
     },
     {
-      "epoch": 18.73,
-      "eval_f1": 0.9312977099236641,
-      "eval_loss": 0.5978376269340515,
-      "eval_runtime": 0.4536,
-      "eval_samples_per_second": 180.765,
-      "eval_steps_per_second": 6.613,
-      "step": 38
+      "epoch": 29.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4697326123714447,
+      "eval_runtime": 0.6728,
+      "eval_samples_per_second": 141.199,
+      "eval_steps_per_second": 4.459,
+      "step": 87
     },
     {
-      "epoch": 19.73,
-      "eval_f1": 0.9312977099236641,
-      "eval_loss": 0.5974743962287903,
-      "eval_runtime": 0.4555,
-      "eval_samples_per_second": 180.031,
-      "eval_steps_per_second": 6.586,
-      "step": 40
+      "epoch": 30.0,
+      "eval_f1": 0.8837209302325582,
+      "eval_loss": 0.4695647060871124,
+      "eval_runtime": 0.6801,
+      "eval_samples_per_second": 139.689,
+      "eval_steps_per_second": 4.411,
+      "step": 90
     },
     {
-      "epoch": 19.73,
-      "step": 40,
-      "total_flos": 1.620732679243776e+17,
-      "train_loss": 0.8578881025314331,
-      "train_runtime": 102.9255,
-      "train_samples_per_second": 63.347,
-      "train_steps_per_second": 0.389
+      "epoch": 30.0,
+      "step": 90,
+      "total_flos": 2.8570124903878656e+17,
+      "train_loss": 0.5013438595665826,
+      "train_runtime": 222.831,
+      "train_samples_per_second": 51.025,
+      "train_steps_per_second": 0.404
     }
   ],
-  "max_steps": 40,
-  "num_train_epochs": 20,
-  "total_flos": 1.620732679243776e+17,
+  "max_steps": 90,
+  "num_train_epochs": 30,
+  "total_flos": 2.8570124903878656e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98064b8c1615330c92e0bd8f4b34a86ff482df6adfeb93cfdfa4d377fa744a73
+oid sha256:6605b4550676653f6a9e7bbcb528bc7fd05170d1b839094df3e2a84195ed6b46
 size 3311