Training in progress, epoch 1

all_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.794973544973545,
-    "eval_f1": 0.7911822260194087,
-    "eval_loss": 0.8158386945724487,
-    "eval_precision": 0.7898693796510783,
-    "eval_recall": 0.794973544973545,
-    "eval_runtime": 8.8626,
-    "eval_samples_per_second": 170.604,
-    "eval_steps_per_second": 10.719,
-    "total_flos": 4.65670232933972e+18,
-    "train_loss": 0.2716429328141264,
-    "train_runtime": 832.7663,
-    "train_samples_per_second": 48.105,
-    "train_steps_per_second": 3.007
+    "eval_accuracy": 0.7817460317460317,
+    "eval_f1": 0.7722244420145986,
+    "eval_loss": 0.8502413630485535,
+    "eval_precision": 0.7756312792366583,
+    "eval_recall": 0.7817460317460317,
+    "eval_runtime": 10.0761,
+    "eval_samples_per_second": 150.058,
+    "eval_steps_per_second": 18.757,
+    "total_flos": 3.104468219559813e+18,
+    "train_loss": 0.0725347773061683,
+    "train_runtime": 849.1916,
+    "train_samples_per_second": 47.174,
+    "train_steps_per_second": 2.949
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.794973544973545,
-    "eval_f1": 0.7911822260194087,
-    "eval_loss": 0.8158386945724487,
-    "eval_precision": 0.7898693796510783,
-    "eval_recall": 0.794973544973545,
-    "eval_runtime": 8.8626,
-    "eval_samples_per_second": 170.604,
-    "eval_steps_per_second": 10.719
+    "epoch": 4.0,
+    "eval_accuracy": 0.7817460317460317,
+    "eval_f1": 0.7722244420145986,
+    "eval_loss": 0.8502413630485535,
+    "eval_precision": 0.7756312792366583,
+    "eval_recall": 0.7817460317460317,
+    "eval_runtime": 10.0761,
+    "eval_samples_per_second": 150.058,
+    "eval_steps_per_second": 18.757
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19b65157a312222f7d90a2baab38d283d36066ea201f5d41216c9e1c16bf7e11
+oid sha256:395c6a4d60ddec8cc483a862b93b051769a3190c99b76fb9739e6c5d32e6955d
 size 343284077

runs/Oct27_21-22-14_EE4E077/events.out.tfevents.1698413879.EE4E077.103115.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:beabc4f6f14a0cd7ef3beefd317bba1f6146e2ad660f7c8a8e2f404faa9bf4d2
+size 560

runs/Oct27_22-15-51_EE4E077/1698416154.775236/events.out.tfevents.1698416154.EE4E077.103115.10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:339b203dbaf2081c885219d6e4a9d823431b0f8e2b4070eb9fba16b8c4653a9b
+size 5919

runs/Oct27_22-15-51_EE4E077/events.out.tfevents.1698416154.EE4E077.103115.9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07476ca587eceeb26e2fe72fb228d090cb8b331fd72311fca7a43997ff57f831
+size 6598

train_results.json

@@ -1,7 +1,8 @@
 {
     "epoch": 4.0,
-    "train_loss": 0.2716429328141264,
-    "train_runtime": 832.7663,
-    "train_samples_per_second": 48.105,
-    "train_steps_per_second": 3.007
+    "total_flos": 3.104468219559813e+18,
+    "train_loss": 0.0725347773061683,
+    "train_runtime": 849.1916,
+    "train_samples_per_second": 47.174,
+    "train_steps_per_second": 2.949
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8860103626943006,
-  "best_model_checkpoint": "./vit-focal-skin/checkpoint-1252",
+  "best_metric": 0.8704663212435233,
+  "best_model_checkpoint": "./vit-focal-skin/checkpoint-1878",
   "epoch": 4.0,
   "global_step": 2504,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
     {
       "epoch": 0.16,
       "learning_rate": 0.00019201277955271565,
-      "loss": 0.7905,
+      "loss": 0.1663,
       "step": 100
     },
     {
       "epoch": 0.32,
       "learning_rate": 0.00018402555910543132,
-      "loss": 0.6162,
+      "loss": 0.1962,
       "step": 200
     },
     {
       "epoch": 0.48,
       "learning_rate": 0.000176038338658147,
-      "loss": 0.5709,
+      "loss": 0.1853,
       "step": 300
     },
     {
       "epoch": 0.64,
       "learning_rate": 0.00016805111821086263,
-      "loss": 0.5101,
+      "loss": 0.1706,
       "step": 400
     },
     {
       "epoch": 0.8,
       "learning_rate": 0.0001600638977635783,
-      "loss": 0.4966,
+      "loss": 0.1593,
       "step": 500
     },
     {
       "epoch": 0.96,
       "learning_rate": 0.00015207667731629394,
-      "loss": 0.4563,
+      "loss": 0.1702,
       "step": 600
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8341968911917098,
-      "eval_f1": 0.8331282144797781,
-      "eval_loss": 0.40268635749816895,
-      "eval_precision": 0.8791963514680484,
-      "eval_recall": 0.8341968911917098,
-      "eval_runtime": 1.5386,
-      "eval_samples_per_second": 125.44,
-      "eval_steps_per_second": 16.249,
+      "eval_accuracy": 0.8393782383419689,
+      "eval_f1": 0.8380951982999452,
+      "eval_loss": 0.39222732186317444,
+      "eval_precision": 0.857762154943502,
+      "eval_recall": 0.8393782383419689,
+      "eval_runtime": 1.5181,
+      "eval_samples_per_second": 127.129,
+      "eval_steps_per_second": 16.467,
       "step": 626
     },
     {
       "epoch": 1.12,
       "learning_rate": 0.00014408945686900958,
-      "loss": 0.385,
+      "loss": 0.1036,
       "step": 700
     },
     {
       "epoch": 1.28,
       "learning_rate": 0.00013610223642172525,
-      "loss": 0.3621,
+      "loss": 0.0941,
       "step": 800
     },
     {
       "epoch": 1.44,
-      "learning_rate": 0.00012811501597444092,
-      "loss": 0.3624,
+      "learning_rate": 0.00012819488817891373,
+      "loss": 0.1029,
       "step": 900
     },
     {
       "epoch": 1.6,
-      "learning_rate": 0.00012012779552715656,
-      "loss": 0.3403,
+      "learning_rate": 0.0001202076677316294,
+      "loss": 0.095,
       "step": 1000
     },
     {
       "epoch": 1.76,
-      "learning_rate": 0.00011214057507987221,
-      "loss": 0.3293,
+      "learning_rate": 0.00011222044728434504,
+      "loss": 0.0911,
       "step": 1100
     },
     {
       "epoch": 1.92,
-      "learning_rate": 0.00010415335463258787,
-      "loss": 0.2954,
+      "learning_rate": 0.00010423322683706072,
+      "loss": 0.0647,
       "step": 1200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8860103626943006,
-      "eval_f1": 0.8891954516071332,
-      "eval_loss": 0.3063604235649109,
-      "eval_precision": 0.8988400162775354,
-      "eval_recall": 0.8860103626943006,
-      "eval_runtime": 1.5206,
-      "eval_samples_per_second": 126.921,
-      "eval_steps_per_second": 16.441,
+      "eval_accuracy": 0.8238341968911918,
+      "eval_f1": 0.8247834676883651,
+      "eval_loss": 0.561523973941803,
+      "eval_precision": 0.8404391615022521,
+      "eval_recall": 0.8238341968911918,
+      "eval_runtime": 1.446,
+      "eval_samples_per_second": 133.47,
+      "eval_steps_per_second": 17.289,
       "step": 1252
     },
     {
       "epoch": 2.08,
-      "learning_rate": 9.616613418530351e-05,
-      "loss": 0.2258,
+      "learning_rate": 9.624600638977636e-05,
+      "loss": 0.052,
       "step": 1300
     },
     {
       "epoch": 2.24,
-      "learning_rate": 8.817891373801918e-05,
-      "loss": 0.1583,
+      "learning_rate": 8.825878594249202e-05,
+      "loss": 0.0298,
       "step": 1400
     },
     {
       "epoch": 2.4,
-      "learning_rate": 8.019169329073483e-05,
-      "loss": 0.1589,
+      "learning_rate": 8.027156549520767e-05,
+      "loss": 0.0243,
       "step": 1500
     },
     {
       "epoch": 2.56,
-      "learning_rate": 7.220447284345049e-05,
-      "loss": 0.1271,
+      "learning_rate": 7.228434504792333e-05,
+      "loss": 0.0332,
       "step": 1600
     },
     {
       "epoch": 2.72,
-      "learning_rate": 6.421725239616614e-05,
-      "loss": 0.1746,
+      "learning_rate": 6.429712460063898e-05,
+      "loss": 0.0267,
       "step": 1700
     },
     {
       "epoch": 2.88,
-      "learning_rate": 5.623003194888179e-05,
-      "loss": 0.1118,
+      "learning_rate": 5.630990415335463e-05,
+      "loss": 0.0111,
       "step": 1800
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8704663212435233,
-      "eval_f1": 0.8663258953141536,
-      "eval_loss": 0.30049628019332886,
-      "eval_precision": 0.8698602729520133,
+      "eval_f1": 0.868408303886561,
+      "eval_loss": 0.4315575659275055,
+      "eval_precision": 0.8670271155479244,
       "eval_recall": 0.8704663212435233,
-      "eval_runtime": 1.4368,
-      "eval_samples_per_second": 134.33,
-      "eval_steps_per_second": 17.4,
+      "eval_runtime": 1.4834,
+      "eval_samples_per_second": 130.104,
+      "eval_steps_per_second": 16.853,
       "step": 1878
     },
     {
       "epoch": 3.04,
-      "learning_rate": 4.824281150159744e-05,
-      "loss": 0.1193,
+      "learning_rate": 4.832268370607029e-05,
+      "loss": 0.0175,
       "step": 1900
     },
     {
       "epoch": 3.19,
-      "learning_rate": 4.0255591054313104e-05,
-      "loss": 0.0409,
+      "learning_rate": 4.0335463258785946e-05,
+      "loss": 0.0083,
       "step": 2000
     },
     {
       "epoch": 3.35,
-      "learning_rate": 3.226837060702875e-05,
-      "loss": 0.0327,
+      "learning_rate": 3.23482428115016e-05,
+      "loss": 0.0038,
       "step": 2100
     },
     {
       "epoch": 3.51,
-      "learning_rate": 2.428115015974441e-05,
-      "loss": 0.0357,
+      "learning_rate": 2.4361022364217255e-05,
+      "loss": 0.0009,
       "step": 2200
     },
     {
       "epoch": 3.67,
-      "learning_rate": 1.6293929712460065e-05,
-      "loss": 0.0386,
+      "learning_rate": 1.6373801916932906e-05,
+      "loss": 0.0034,
       "step": 2300
     },
     {
       "epoch": 3.83,
-      "learning_rate": 8.306709265175718e-06,
-      "loss": 0.0303,
+      "learning_rate": 8.386581469648563e-06,
+      "loss": 0.0025,
       "step": 2400
     },
     {
       "epoch": 3.99,
-      "learning_rate": 3.194888178913738e-07,
-      "loss": 0.0317,
+      "learning_rate": 3.9936102236421723e-07,
+      "loss": 0.0034,
       "step": 2500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8549222797927462,
-      "eval_f1": 0.8560478324319817,
-      "eval_loss": 0.3552953004837036,
-      "eval_precision": 0.8595484172497833,
-      "eval_recall": 0.8549222797927462,
-      "eval_runtime": 1.5113,
-      "eval_samples_per_second": 127.703,
-      "eval_steps_per_second": 16.542,
+      "eval_accuracy": 0.8601036269430051,
+      "eval_f1": 0.8616514102008719,
+      "eval_loss": 0.4513249099254608,
+      "eval_precision": 0.8650015326151586,
+      "eval_recall": 0.8601036269430051,
+      "eval_runtime": 1.498,
+      "eval_samples_per_second": 128.838,
+      "eval_steps_per_second": 16.689,
       "step": 2504
     },
     {
       "epoch": 4.0,
       "step": 2504,
       "total_flos": 3.104468219559813e+18,
-      "train_loss": 0.2716429328141264,
-      "train_runtime": 832.7663,
-      "train_samples_per_second": 48.105,
-      "train_steps_per_second": 3.007
+      "train_loss": 0.0725347773061683,
+      "train_runtime": 849.1916,
+      "train_samples_per_second": 47.174,
+      "train_steps_per_second": 2.949
     }
   ],
   "max_steps": 2504,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dae95784d5a6901f55f2c792e11ade68b7f5871a7f402bd5113f06400a3ea472
+oid sha256:3f657e798e7f77936234299466c6296e6c9fba3c3d1d364c041c99c8e3780c6a
 size 3899