End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 4.98,
-    "eval_accuracy": 0.9955262036642523,
-    "eval_loss": 0.014164051972329617,
-    "eval_runtime": 156.8919,
-    "eval_samples_per_second": 29.919,
-    "eval_steps_per_second": 0.937,
-    "total_flos": 1.8120726851476193e+18,
-    "train_loss": 0.08313048457105955,
-    "train_runtime": 2165.5413,
-    "train_samples_per_second": 10.838,
-    "train_steps_per_second": 0.083
+    "epoch": 2.98,
+    "eval_accuracy": 0.989987217724755,
+    "eval_loss": 0.03180089592933655,
+    "eval_runtime": 157.5304,
+    "eval_samples_per_second": 29.797,
+    "eval_steps_per_second": 0.933,
+    "total_flos": 1.0845778866458296e+18,
+    "train_loss": 0.11129519205402445,
+    "train_runtime": 1326.3431,
+    "train_samples_per_second": 10.617,
+    "train_steps_per_second": 0.081
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.98,
-    "eval_accuracy": 0.9955262036642523,
-    "eval_loss": 0.014164051972329617,
-    "eval_runtime": 156.8919,
-    "eval_samples_per_second": 29.919,
-    "eval_steps_per_second": 0.937
+    "epoch": 2.98,
+    "eval_accuracy": 0.989987217724755,
+    "eval_loss": 0.03180089592933655,
+    "eval_runtime": 157.5304,
+    "eval_samples_per_second": 29.797,
+    "eval_steps_per_second": 0.933
 }

runs/Feb05_23-22-29_910d9832454d/events.out.tfevents.1675640939.910d9832454d.294.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e871e516b456d5916d8aa874473d75d705c7d0f9469be6f779f71cfee6ab627b
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.98,
-    "total_flos": 1.8120726851476193e+18,
-    "train_loss": 0.08313048457105955,
-    "train_runtime": 2165.5413,
-    "train_samples_per_second": 10.838,
-    "train_steps_per_second": 0.083
+    "epoch": 2.98,
+    "total_flos": 1.0845778866458296e+18,
+    "train_loss": 0.11129519205402445,
+    "train_runtime": 1326.3431,
+    "train_samples_per_second": 10.617,
+    "train_steps_per_second": 0.081
 }

trainer_state.json

@@ -1,286 +1,178 @@
 {
-  "best_metric": 0.9957392415850022,
-  "best_model_checkpoint": "vit-base-patch16-224-finetuned-chest/checkpoint-180",
-  "epoch": 4.979591836734694,
-  "global_step": 180,
+  "best_metric": 0.9887089902002556,
+  "best_model_checkpoint": "vit-base-patch16-224-finetuned-chest/checkpoint-108",
+  "epoch": 2.979591836734694,
+  "global_step": 108,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.14,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.6592,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.5974,
       "step": 5
     },
     {
       "epoch": 0.27,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.4472,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 0.3824,
       "step": 10
     },
     {
       "epoch": 0.41,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.2618,
+      "learning_rate": 4.793814432989691e-05,
+      "loss": 0.1834,
       "step": 15
     },
     {
       "epoch": 0.54,
-      "learning_rate": 4.938271604938271e-05,
-      "loss": 0.1677,
+      "learning_rate": 4.536082474226804e-05,
+      "loss": 0.1151,
       "step": 20
     },
     {
       "epoch": 0.68,
-      "learning_rate": 4.783950617283951e-05,
-      "loss": 0.1731,
+      "learning_rate": 4.2783505154639175e-05,
+      "loss": 0.1198,
       "step": 25
     },
     {
       "epoch": 0.82,
-      "learning_rate": 4.62962962962963e-05,
-      "loss": 0.0952,
+      "learning_rate": 4.020618556701031e-05,
+      "loss": 0.1117,
       "step": 30
     },
     {
       "epoch": 0.95,
-      "learning_rate": 4.4753086419753084e-05,
-      "loss": 0.1288,
+      "learning_rate": 3.7628865979381445e-05,
+      "loss": 0.0947,
       "step": 35
     },
     {
       "epoch": 0.98,
-      "eval_accuracy": 0.9527055815935237,
-      "eval_loss": 0.10909163951873779,
-      "eval_runtime": 152.5532,
-      "eval_samples_per_second": 30.77,
-      "eval_steps_per_second": 0.964,
+      "eval_accuracy": 0.9731572219855135,
+      "eval_loss": 0.07849641144275665,
+      "eval_runtime": 153.1267,
+      "eval_samples_per_second": 30.654,
+      "eval_steps_per_second": 0.96,
       "step": 36
     },
     {
       "epoch": 1.11,
-      "learning_rate": 4.3209876543209875e-05,
-      "loss": 0.1009,
+      "learning_rate": 3.5051546391752576e-05,
+      "loss": 0.0707,
       "step": 40
     },
     {
       "epoch": 1.24,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.0668,
+      "learning_rate": 3.2474226804123714e-05,
+      "loss": 0.0719,
       "step": 45
     },
     {
       "epoch": 1.38,
-      "learning_rate": 4.012345679012346e-05,
-      "loss": 0.0414,
+      "learning_rate": 2.9896907216494846e-05,
+      "loss": 0.0655,
       "step": 50
     },
     {
       "epoch": 1.52,
-      "learning_rate": 3.8580246913580246e-05,
-      "loss": 0.0553,
+      "learning_rate": 2.7319587628865977e-05,
+      "loss": 0.0661,
       "step": 55
     },
     {
       "epoch": 1.65,
-      "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.074,
+      "learning_rate": 2.4742268041237116e-05,
+      "loss": 0.0557,
       "step": 60
     },
     {
       "epoch": 1.79,
-      "learning_rate": 3.5493827160493834e-05,
-      "loss": 0.0478,
+      "learning_rate": 2.2164948453608247e-05,
+      "loss": 0.0707,
       "step": 65
     },
     {
       "epoch": 1.93,
-      "learning_rate": 3.395061728395062e-05,
-      "loss": 0.0458,
+      "learning_rate": 1.9587628865979382e-05,
+      "loss": 0.048,
       "step": 70
     },
     {
       "epoch": 1.98,
-      "eval_accuracy": 0.9865786109927567,
-      "eval_loss": 0.040942613035440445,
-      "eval_runtime": 153.8543,
-      "eval_samples_per_second": 30.509,
-      "eval_steps_per_second": 0.955,
+      "eval_accuracy": 0.9731572219855135,
+      "eval_loss": 0.06775445491075516,
+      "eval_runtime": 154.5708,
+      "eval_samples_per_second": 30.368,
+      "eval_steps_per_second": 0.951,
       "step": 72
     },
     {
       "epoch": 2.08,
-      "learning_rate": 3.240740740740741e-05,
-      "loss": 0.0601,
+      "learning_rate": 1.7010309278350517e-05,
+      "loss": 0.0682,
       "step": 75
     },
     {
       "epoch": 2.22,
-      "learning_rate": 3.08641975308642e-05,
-      "loss": 0.0508,
+      "learning_rate": 1.4432989690721649e-05,
+      "loss": 0.0564,
       "step": 80
     },
     {
       "epoch": 2.35,
-      "learning_rate": 2.9320987654320992e-05,
-      "loss": 0.0335,
+      "learning_rate": 1.1855670103092783e-05,
+      "loss": 0.0499,
       "step": 85
     },
     {
       "epoch": 2.49,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.0443,
+      "learning_rate": 9.278350515463918e-06,
+      "loss": 0.0412,
       "step": 90
     },
     {
       "epoch": 2.63,
-      "learning_rate": 2.623456790123457e-05,
-      "loss": 0.0218,
+      "learning_rate": 6.701030927835052e-06,
+      "loss": 0.045,
       "step": 95
     },
     {
       "epoch": 2.76,
-      "learning_rate": 2.4691358024691357e-05,
-      "loss": 0.0327,
+      "learning_rate": 4.123711340206186e-06,
+      "loss": 0.0353,
       "step": 100
     },
     {
       "epoch": 2.9,
-      "learning_rate": 2.314814814814815e-05,
-      "loss": 0.0488,
+      "learning_rate": 1.5463917525773197e-06,
+      "loss": 0.0352,
       "step": 105
     },
     {
       "epoch": 2.98,
-      "eval_accuracy": 0.9906263314870047,
-      "eval_loss": 0.023438777774572372,
-      "eval_runtime": 155.1826,
-      "eval_samples_per_second": 30.248,
-      "eval_steps_per_second": 0.947,
+      "eval_accuracy": 0.9887089902002556,
+      "eval_loss": 0.03288055956363678,
+      "eval_runtime": 156.1442,
+      "eval_samples_per_second": 30.062,
+      "eval_steps_per_second": 0.941,
       "step": 108
     },
     {
-      "epoch": 3.05,
-      "learning_rate": 2.1604938271604937e-05,
-      "loss": 0.0409,
-      "step": 110
-    },
-    {
-      "epoch": 3.19,
-      "learning_rate": 2.006172839506173e-05,
-      "loss": 0.0248,
-      "step": 115
-    },
-    {
-      "epoch": 3.33,
-      "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.0259,
-      "step": 120
-    },
-    {
-      "epoch": 3.46,
-      "learning_rate": 1.697530864197531e-05,
-      "loss": 0.0165,
-      "step": 125
-    },
-    {
-      "epoch": 3.6,
-      "learning_rate": 1.54320987654321e-05,
-      "loss": 0.0324,
-      "step": 130
-    },
-    {
-      "epoch": 3.73,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.0314,
-      "step": 135
-    },
-    {
-      "epoch": 3.87,
-      "learning_rate": 1.2345679012345678e-05,
-      "loss": 0.0287,
-      "step": 140
-    },
-    {
-      "epoch": 3.98,
-      "eval_accuracy": 0.994034938219003,
-      "eval_loss": 0.018356822431087494,
-      "eval_runtime": 154.3825,
-      "eval_samples_per_second": 30.405,
-      "eval_steps_per_second": 0.952,
-      "step": 144
-    },
-    {
-      "epoch": 4.03,
-      "learning_rate": 1.0802469135802469e-05,
-      "loss": 0.0336,
-      "step": 145
-    },
-    {
-      "epoch": 4.16,
-      "learning_rate": 9.259259259259259e-06,
-      "loss": 0.0191,
-      "step": 150
-    },
-    {
-      "epoch": 4.3,
-      "learning_rate": 7.71604938271605e-06,
-      "loss": 0.0161,
-      "step": 155
-    },
-    {
-      "epoch": 4.44,
-      "learning_rate": 6.172839506172839e-06,
-      "loss": 0.0096,
-      "step": 160
-    },
-    {
-      "epoch": 4.57,
-      "learning_rate": 4.6296296296296296e-06,
-      "loss": 0.0119,
-      "step": 165
-    },
-    {
-      "epoch": 4.71,
-      "learning_rate": 3.0864197530864196e-06,
-      "loss": 0.0184,
-      "step": 170
-    },
-    {
-      "epoch": 4.84,
-      "learning_rate": 1.5432098765432098e-06,
-      "loss": 0.0091,
-      "step": 175
-    },
-    {
-      "epoch": 4.98,
-      "learning_rate": 0.0,
-      "loss": 0.0174,
-      "step": 180
-    },
-    {
-      "epoch": 4.98,
-      "eval_accuracy": 0.9957392415850022,
-      "eval_loss": 0.013076434843242168,
-      "eval_runtime": 153.7061,
-      "eval_samples_per_second": 30.539,
-      "eval_steps_per_second": 0.956,
-      "step": 180
-    },
-    {
-      "epoch": 4.98,
-      "step": 180,
-      "total_flos": 1.8120726851476193e+18,
-      "train_loss": 0.08313048457105955,
-      "train_runtime": 2165.5413,
-      "train_samples_per_second": 10.838,
-      "train_steps_per_second": 0.083
+      "epoch": 2.98,
+      "step": 108,
+      "total_flos": 1.0845778866458296e+18,
+      "train_loss": 0.11129519205402445,
+      "train_runtime": 1326.3431,
+      "train_samples_per_second": 10.617,
+      "train_steps_per_second": 0.081
     }
   ],
-  "max_steps": 180,
-  "num_train_epochs": 5,
-  "total_flos": 1.8120726851476193e+18,
+  "max_steps": 108,
+  "num_train_epochs": 3,
+  "total_flos": 1.0845778866458296e+18,
   "trial_name": null,
   "trial_params": null
 }