End of training

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 3.99,
-    "eval_accuracy": 0.9613120269133726,
-    "eval_f1": 0.9613120269133726,
-    "eval_loss": 0.1092575266957283,
-    "eval_precision": 0.9613120269133726,
-    "eval_recall": 0.9613120269133726,
-    "eval_runtime": 58.2407,
-    "eval_samples_per_second": 20.415,
-    "eval_steps_per_second": 0.652,
-    "total_flos": 1.061932048026071e+18,
-    "train_loss": 0.4249281358108463,
-    "train_runtime": 6173.988,
-    "train_samples_per_second": 6.932,
-    "train_steps_per_second": 0.054
+    "epoch": 4.99,
+    "eval_accuracy": 0.9646761984861227,
+    "eval_f1": 0.9646761984861227,
+    "eval_loss": 0.10117336362600327,
+    "eval_precision": 0.9646761984861227,
+    "eval_recall": 0.9646761984861227,
+    "eval_runtime": 60.8804,
+    "eval_samples_per_second": 19.53,
+    "eval_steps_per_second": 0.624,
+    "total_flos": 1.3278811358711808e+18,
+    "train_loss": 0.383831198531461,
+    "train_runtime": 7560.7227,
+    "train_samples_per_second": 7.075,
+    "train_steps_per_second": 0.055
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 3.99,
-    "eval_accuracy": 0.9613120269133726,
-    "eval_f1": 0.9613120269133726,
-    "eval_loss": 0.1092575266957283,
-    "eval_precision": 0.9613120269133726,
-    "eval_recall": 0.9613120269133726,
-    "eval_runtime": 58.2407,
-    "eval_samples_per_second": 20.415,
-    "eval_steps_per_second": 0.652
+    "epoch": 4.99,
+    "eval_accuracy": 0.9646761984861227,
+    "eval_f1": 0.9646761984861227,
+    "eval_loss": 0.10117336362600327,
+    "eval_precision": 0.9646761984861227,
+    "eval_recall": 0.9646761984861227,
+    "eval_runtime": 60.8804,
+    "eval_samples_per_second": 19.53,
+    "eval_steps_per_second": 0.624
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.99,
-    "total_flos": 1.061932048026071e+18,
-    "train_loss": 0.4249281358108463,
-    "train_runtime": 6173.988,
-    "train_samples_per_second": 6.932,
-    "train_steps_per_second": 0.054
+    "epoch": 4.99,
+    "total_flos": 1.3278811358711808e+18,
+    "train_loss": 0.383831198531461,
+    "train_runtime": 7560.7227,
+    "train_samples_per_second": 7.075,
+    "train_steps_per_second": 0.055
 }

trainer_state.json

@@ -1,271 +1,331 @@
 {
-  "best_metric": 0.9613120269133726,
-  "best_model_checkpoint": "Brain_Tumor_Classification/checkpoint-332",
-  "epoch": 3.991044776119403,
-  "global_step": 332,
+  "best_metric": 0.9646761984861227,
+  "best_model_checkpoint": "Brain_Tumor_Classification/checkpoint-415",
+  "epoch": 4.991044776119403,
+  "global_step": 415,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.12,
-      "learning_rate": 1.4705882352941177e-05,
-      "loss": 1.4267,
+      "learning_rate": 1.1904761904761905e-05,
+      "loss": 1.3844,
       "step": 10
     },
     {
       "epoch": 0.24,
-      "learning_rate": 2.9411764705882354e-05,
-      "loss": 1.2467,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 1.2375,
       "step": 20
     },
     {
       "epoch": 0.36,
-      "learning_rate": 4.411764705882353e-05,
-      "loss": 1.0154,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.9742,
       "step": 30
     },
     {
       "epoch": 0.48,
-      "learning_rate": 4.8993288590604034e-05,
-      "loss": 0.709,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 0.7601,
       "step": 40
     },
     {
       "epoch": 0.6,
-      "learning_rate": 4.731543624161074e-05,
-      "loss": 0.604,
+      "learning_rate": 4.8927613941018765e-05,
+      "loss": 0.7047,
       "step": 50
     },
     {
       "epoch": 0.72,
-      "learning_rate": 4.5637583892617453e-05,
-      "loss": 0.5955,
+      "learning_rate": 4.7587131367292224e-05,
+      "loss": 0.5255,
       "step": 60
     },
     {
       "epoch": 0.84,
-      "learning_rate": 4.395973154362416e-05,
-      "loss": 0.485,
+      "learning_rate": 4.624664879356568e-05,
+      "loss": 0.5059,
       "step": 70
     },
     {
       "epoch": 0.96,
-      "learning_rate": 4.228187919463087e-05,
-      "loss": 0.4369,
+      "learning_rate": 4.490616621983915e-05,
+      "loss": 0.4856,
       "step": 80
     },
     {
       "epoch": 0.99,
-      "eval_accuracy": 0.9091673675357443,
-      "eval_f1": 0.9091673675357443,
-      "eval_loss": 0.2500155568122864,
-      "eval_precision": 0.9091673675357443,
-      "eval_recall": 0.9091673675357443,
-      "eval_runtime": 64.8347,
-      "eval_samples_per_second": 18.339,
-      "eval_steps_per_second": 0.586,
+      "eval_accuracy": 0.8444070647603028,
+      "eval_f1": 0.8444070647603028,
+      "eval_loss": 0.3771306276321411,
+      "eval_precision": 0.8444070647603028,
+      "eval_recall": 0.8444070647603028,
+      "eval_runtime": 65.7487,
+      "eval_samples_per_second": 18.084,
+      "eval_steps_per_second": 0.578,
       "step": 83
     },
     {
       "epoch": 1.08,
-      "learning_rate": 4.060402684563759e-05,
-      "loss": 0.4492,
+      "learning_rate": 4.35656836461126e-05,
+      "loss": 0.4626,
       "step": 90
     },
     {
       "epoch": 1.2,
-      "learning_rate": 3.89261744966443e-05,
-      "loss": 0.3878,
+      "learning_rate": 4.222520107238606e-05,
+      "loss": 0.4249,
       "step": 100
     },
     {
       "epoch": 1.32,
-      "learning_rate": 3.724832214765101e-05,
-      "loss": 0.3951,
+      "learning_rate": 4.088471849865952e-05,
+      "loss": 0.3855,
       "step": 110
     },
     {
       "epoch": 1.44,
-      "learning_rate": 3.557046979865772e-05,
-      "loss": 0.3708,
+      "learning_rate": 3.954423592493298e-05,
+      "loss": 0.385,
       "step": 120
     },
     {
       "epoch": 1.56,
-      "learning_rate": 3.389261744966443e-05,
-      "loss": 0.342,
+      "learning_rate": 3.8203753351206436e-05,
+      "loss": 0.3522,
       "step": 130
     },
     {
       "epoch": 1.68,
-      "learning_rate": 3.221476510067114e-05,
-      "loss": 0.3343,
+      "learning_rate": 3.6863270777479895e-05,
+      "loss": 0.367,
       "step": 140
     },
     {
       "epoch": 1.8,
-      "learning_rate": 3.0536912751677856e-05,
-      "loss": 0.314,
+      "learning_rate": 3.5522788203753354e-05,
+      "loss": 0.3172,
       "step": 150
     },
     {
       "epoch": 1.92,
-      "learning_rate": 2.885906040268457e-05,
-      "loss": 0.3777,
+      "learning_rate": 3.418230563002681e-05,
+      "loss": 0.3495,
       "step": 160
     },
     {
       "epoch": 1.99,
-      "eval_accuracy": 0.9301934398654331,
-      "eval_f1": 0.9301934398654331,
-      "eval_loss": 0.17629508674144745,
-      "eval_precision": 0.9301934398654331,
-      "eval_recall": 0.9301934398654331,
-      "eval_runtime": 69.6074,
-      "eval_samples_per_second": 17.082,
-      "eval_steps_per_second": 0.546,
+      "eval_accuracy": 0.894869638351556,
+      "eval_f1": 0.894869638351556,
+      "eval_loss": 0.2608145773410797,
+      "eval_precision": 0.894869638351556,
+      "eval_recall": 0.894869638351556,
+      "eval_runtime": 66.7411,
+      "eval_samples_per_second": 17.815,
+      "eval_steps_per_second": 0.569,
       "step": 166
     },
     {
       "epoch": 2.05,
-      "learning_rate": 2.7181208053691276e-05,
-      "loss": 0.3736,
+      "learning_rate": 3.284182305630027e-05,
+      "loss": 0.3677,
       "step": 170
     },
     {
       "epoch": 2.17,
-      "learning_rate": 2.550335570469799e-05,
-      "loss": 0.3141,
+      "learning_rate": 3.1501340482573724e-05,
+      "loss": 0.2809,
       "step": 180
     },
     {
       "epoch": 2.29,
-      "learning_rate": 2.38255033557047e-05,
-      "loss": 0.3334,
+      "learning_rate": 3.0160857908847186e-05,
+      "loss": 0.297,
       "step": 190
     },
     {
       "epoch": 2.41,
-      "learning_rate": 2.2147651006711412e-05,
-      "loss": 0.3104,
+      "learning_rate": 2.8820375335120648e-05,
+      "loss": 0.3134,
       "step": 200
     },
     {
       "epoch": 2.53,
-      "learning_rate": 2.0469798657718122e-05,
-      "loss": 0.266,
+      "learning_rate": 2.7479892761394104e-05,
+      "loss": 0.2306,
       "step": 210
     },
     {
       "epoch": 2.64,
-      "learning_rate": 1.8791946308724835e-05,
-      "loss": 0.2586,
+      "learning_rate": 2.6139410187667563e-05,
+      "loss": 0.2442,
       "step": 220
     },
     {
       "epoch": 2.76,
-      "learning_rate": 1.7114093959731545e-05,
-      "loss": 0.2747,
+      "learning_rate": 2.479892761394102e-05,
+      "loss": 0.2885,
       "step": 230
     },
     {
       "epoch": 2.88,
-      "learning_rate": 1.5436241610738255e-05,
-      "loss": 0.2684,
+      "learning_rate": 2.3458445040214477e-05,
+      "loss": 0.252,
       "step": 240
     },
     {
       "epoch": 2.99,
-      "eval_accuracy": 0.9512195121951219,
-      "eval_f1": 0.9512195121951219,
-      "eval_loss": 0.12147016078233719,
-      "eval_precision": 0.9512195121951219,
-      "eval_recall": 0.9512195121951219,
-      "eval_runtime": 62.5572,
-      "eval_samples_per_second": 19.007,
-      "eval_steps_per_second": 0.607,
+      "eval_accuracy": 0.9486963835155593,
+      "eval_f1": 0.9486963835155594,
+      "eval_loss": 0.14449167251586914,
+      "eval_precision": 0.9486963835155593,
+      "eval_recall": 0.9486963835155593,
+      "eval_runtime": 64.3986,
+      "eval_samples_per_second": 18.463,
+      "eval_steps_per_second": 0.59,
       "step": 249
     },
     {
       "epoch": 3.01,
-      "learning_rate": 1.3758389261744966e-05,
-      "loss": 0.2589,
+      "learning_rate": 2.211796246648794e-05,
+      "loss": 0.2939,
       "step": 250
     },
     {
       "epoch": 3.13,
-      "learning_rate": 1.208053691275168e-05,
-      "loss": 0.2467,
+      "learning_rate": 2.0777479892761395e-05,
+      "loss": 0.2099,
       "step": 260
     },
     {
       "epoch": 3.25,
-      "learning_rate": 1.040268456375839e-05,
-      "loss": 0.2406,
+      "learning_rate": 1.9436997319034853e-05,
+      "loss": 0.2477,
       "step": 270
     },
     {
       "epoch": 3.37,
-      "learning_rate": 8.724832214765101e-06,
-      "loss": 0.2289,
+      "learning_rate": 1.8096514745308312e-05,
+      "loss": 0.2674,
       "step": 280
     },
     {
       "epoch": 3.49,
-      "learning_rate": 7.046979865771812e-06,
-      "loss": 0.2812,
+      "learning_rate": 1.675603217158177e-05,
+      "loss": 0.2538,
       "step": 290
     },
     {
       "epoch": 3.61,
-      "learning_rate": 5.3691275167785235e-06,
-      "loss": 0.2478,
+      "learning_rate": 1.5415549597855227e-05,
+      "loss": 0.2201,
       "step": 300
     },
     {
       "epoch": 3.73,
-      "learning_rate": 3.6912751677852355e-06,
-      "loss": 0.2235,
+      "learning_rate": 1.4075067024128689e-05,
+      "loss": 0.2591,
       "step": 310
     },
     {
       "epoch": 3.85,
-      "learning_rate": 2.013422818791946e-06,
-      "loss": 0.207,
+      "learning_rate": 1.2734584450402146e-05,
+      "loss": 0.2087,
       "step": 320
     },
     {
       "epoch": 3.97,
-      "learning_rate": 3.355704697986577e-07,
-      "loss": 0.2363,
+      "learning_rate": 1.1394101876675605e-05,
+      "loss": 0.2364,
       "step": 330
     },
     {
       "epoch": 3.99,
-      "eval_accuracy": 0.9613120269133726,
-      "eval_f1": 0.9613120269133726,
-      "eval_loss": 0.1092575266957283,
-      "eval_precision": 0.9613120269133726,
-      "eval_recall": 0.9613120269133726,
-      "eval_runtime": 64.3459,
-      "eval_samples_per_second": 18.478,
-      "eval_steps_per_second": 0.591,
+      "eval_accuracy": 0.95878889823381,
+      "eval_f1": 0.95878889823381,
+      "eval_loss": 0.10286473482847214,
+      "eval_precision": 0.95878889823381,
+      "eval_recall": 0.95878889823381,
+      "eval_runtime": 65.1006,
+      "eval_samples_per_second": 18.264,
+      "eval_steps_per_second": 0.584,
       "step": 332
     },
     {
-      "epoch": 3.99,
-      "step": 332,
-      "total_flos": 1.061932048026071e+18,
-      "train_loss": 0.4249281358108463,
-      "train_runtime": 6173.988,
-      "train_samples_per_second": 6.932,
-      "train_steps_per_second": 0.054
+      "epoch": 4.1,
+      "learning_rate": 1.0053619302949062e-05,
+      "loss": 0.2312,
+      "step": 340
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 8.71313672922252e-06,
+      "loss": 0.2011,
+      "step": 350
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 7.372654155495978e-06,
+      "loss": 0.1961,
+      "step": 360
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 6.032171581769437e-06,
+      "loss": 0.2277,
+      "step": 370
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 4.691689008042896e-06,
+      "loss": 0.2084,
+      "step": 380
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.351206434316354e-06,
+      "loss": 0.2478,
+      "step": 390
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 2.0107238605898126e-06,
+      "loss": 0.1992,
+      "step": 400
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 6.702412868632708e-07,
+      "loss": 0.2178,
+      "step": 410
+    },
+    {
+      "epoch": 4.99,
+      "eval_accuracy": 0.9646761984861227,
+      "eval_f1": 0.9646761984861227,
+      "eval_loss": 0.10117336362600327,
+      "eval_precision": 0.9646761984861227,
+      "eval_recall": 0.9646761984861227,
+      "eval_runtime": 66.0046,
+      "eval_samples_per_second": 18.014,
+      "eval_steps_per_second": 0.576,
+      "step": 415
+    },
+    {
+      "epoch": 4.99,
+      "step": 415,
+      "total_flos": 1.3278811358711808e+18,
+      "train_loss": 0.383831198531461,
+      "train_runtime": 7560.7227,
+      "train_samples_per_second": 7.075,
+      "train_steps_per_second": 0.055
     }
   ],
-  "max_steps": 332,
-  "num_train_epochs": 4,
-  "total_flos": 1.061932048026071e+18,
+  "max_steps": 415,
+  "num_train_epochs": 5,
+  "total_flos": 1.3278811358711808e+18,
   "trial_name": null,
   "trial_params": null
 }