Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 15.0,
     "total_flos": 1.16237984421888e+18,
-    "train_loss": 0.3610760450363159,
-    "train_runtime": 974.8105,
-    "train_samples_per_second": 15.388,
-    "train_steps_per_second": 0.062
+    "train_loss": 0.475440772374471,
+    "train_runtime": 1200.7807,
+    "train_samples_per_second": 12.492,
+    "train_steps_per_second": 0.05
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7aff6f29850e45f9d3809502ee06082d45a2c6e2e81451cb035264406def030
+oid sha256:70e0c5f48e4cab99d8880a0d85047bb62b1211338a7363e01dd3f6fc698f564a
 size 343268717

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 15.0,
     "total_flos": 1.16237984421888e+18,
-    "train_loss": 0.3610760450363159,
-    "train_runtime": 974.8105,
-    "train_samples_per_second": 15.388,
-    "train_steps_per_second": 0.062
+    "train_loss": 0.475440772374471,
+    "train_runtime": 1200.7807,
+    "train_samples_per_second": 12.492,
+    "train_steps_per_second": 0.05
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9083333333333333,
-  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-40",
+  "best_metric": 0.7916666666666666,
+  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-52",
   "epoch": 15.0,
   "eval_steps": 500,
   "global_step": 60,
@@ -10,227 +10,227 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8666666666666667,
-      "eval_f1_score": 0.8224224224224224,
-      "eval_loss": 0.5036832094192505,
-      "eval_precision": 0.8150354609929079,
-      "eval_recall": 0.8666666666666667,
-      "eval_runtime": 4.0852,
-      "eval_samples_per_second": 58.748,
-      "eval_steps_per_second": 0.979,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_f1_score": 0.6280486019616454,
+      "eval_loss": 0.6009885668754578,
+      "eval_precision": 0.6725490196078432,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 73.0728,
+      "eval_samples_per_second": 3.284,
+      "eval_steps_per_second": 0.055,
       "step": 4
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8708333333333333,
-      "eval_f1_score": 0.8107089829250185,
-      "eval_loss": 0.35003846883773804,
-      "eval_precision": 0.7583506944444445,
-      "eval_recall": 0.8708333333333333,
-      "eval_runtime": 4.1025,
-      "eval_samples_per_second": 58.501,
-      "eval_steps_per_second": 0.975,
+      "eval_accuracy": 0.7375,
+      "eval_f1_score": 0.6302131603336423,
+      "eval_loss": 0.5551710724830627,
+      "eval_precision": 0.8066945606694561,
+      "eval_recall": 0.7375,
+      "eval_runtime": 4.1084,
+      "eval_samples_per_second": 58.416,
+      "eval_steps_per_second": 0.974,
       "step": 8
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8708333333333333,
-      "eval_f1_score": 0.8107089829250185,
-      "eval_loss": 0.3154493272304535,
-      "eval_precision": 0.7583506944444445,
-      "eval_recall": 0.8708333333333333,
-      "eval_runtime": 3.9805,
-      "eval_samples_per_second": 60.295,
-      "eval_steps_per_second": 1.005,
+      "eval_accuracy": 0.7541666666666667,
+      "eval_f1_score": 0.6781685795384426,
+      "eval_loss": 0.5449685454368591,
+      "eval_precision": 0.7597883597883598,
+      "eval_recall": 0.7541666666666667,
+      "eval_runtime": 4.0131,
+      "eval_samples_per_second": 59.804,
+      "eval_steps_per_second": 0.997,
       "step": 12
     },
     {
       "epoch": 3.75,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 0.5284,
+      "loss": 0.576,
       "step": 15
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8833333333333333,
-      "eval_f1_score": 0.849658648884655,
-      "eval_loss": 0.2973836362361908,
-      "eval_precision": 0.8659207030451664,
-      "eval_recall": 0.8833333333333333,
-      "eval_runtime": 3.9445,
-      "eval_samples_per_second": 60.844,
-      "eval_steps_per_second": 1.014,
+      "eval_accuracy": 0.75,
+      "eval_f1_score": 0.6641114982578398,
+      "eval_loss": 0.5325278043746948,
+      "eval_precision": 0.7706552706552707,
+      "eval_recall": 0.75,
+      "eval_runtime": 4.224,
+      "eval_samples_per_second": 56.818,
+      "eval_steps_per_second": 0.947,
       "step": 16
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8875,
-      "eval_f1_score": 0.8768099547511312,
-      "eval_loss": 0.2953941226005554,
-      "eval_precision": 0.873125,
-      "eval_recall": 0.8875,
-      "eval_runtime": 4.0614,
-      "eval_samples_per_second": 59.092,
-      "eval_steps_per_second": 0.985,
+      "eval_accuracy": 0.75,
+      "eval_f1_score": 0.6899999999999998,
+      "eval_loss": 0.5234411954879761,
+      "eval_precision": 0.7232142857142857,
+      "eval_recall": 0.75,
+      "eval_runtime": 3.9713,
+      "eval_samples_per_second": 60.434,
+      "eval_steps_per_second": 1.007,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8958333333333334,
-      "eval_f1_score": 0.8716479619238106,
-      "eval_loss": 0.2720634639263153,
-      "eval_precision": 0.8870833333333333,
-      "eval_recall": 0.8958333333333334,
-      "eval_runtime": 4.0196,
-      "eval_samples_per_second": 59.708,
-      "eval_steps_per_second": 0.995,
+      "eval_accuracy": 0.7625,
+      "eval_f1_score": 0.7075837742504409,
+      "eval_loss": 0.511196494102478,
+      "eval_precision": 0.7505671326826696,
+      "eval_recall": 0.7625,
+      "eval_runtime": 3.9183,
+      "eval_samples_per_second": 61.251,
+      "eval_steps_per_second": 1.021,
       "step": 24
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8875,
-      "eval_f1_score": 0.852667622475749,
-      "eval_loss": 0.26786690950393677,
-      "eval_precision": 0.881712962962963,
-      "eval_recall": 0.8875,
-      "eval_runtime": 4.2691,
-      "eval_samples_per_second": 56.218,
-      "eval_steps_per_second": 0.937,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_f1_score": 0.7221264707039705,
+      "eval_loss": 0.5081896185874939,
+      "eval_precision": 0.7503197108701696,
+      "eval_recall": 0.7666666666666667,
+      "eval_runtime": 4.0209,
+      "eval_samples_per_second": 59.688,
+      "eval_steps_per_second": 0.995,
       "step": 28
     },
     {
       "epoch": 7.5,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.3362,
+      "loss": 0.4876,
       "step": 30
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8875,
-      "eval_f1_score": 0.852667622475749,
-      "eval_loss": 0.26340213418006897,
-      "eval_precision": 0.881712962962963,
-      "eval_recall": 0.8875,
-      "eval_runtime": 4.2103,
-      "eval_samples_per_second": 57.002,
-      "eval_steps_per_second": 0.95,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_f1_score": 0.7287749287749288,
+      "eval_loss": 0.5066744089126587,
+      "eval_precision": 0.746633117661155,
+      "eval_recall": 0.7666666666666667,
+      "eval_runtime": 3.9922,
+      "eval_samples_per_second": 60.118,
+      "eval_steps_per_second": 1.002,
       "step": 32
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9041666666666667,
-      "eval_f1_score": 0.8879374201787995,
-      "eval_loss": 0.25068628787994385,
-      "eval_precision": 0.8952565318162663,
-      "eval_recall": 0.9041666666666667,
-      "eval_runtime": 3.9517,
-      "eval_samples_per_second": 60.733,
-      "eval_steps_per_second": 1.012,
+      "eval_accuracy": 0.7791666666666667,
+      "eval_f1_score": 0.7528159852134981,
+      "eval_loss": 0.5091281533241272,
+      "eval_precision": 0.76231884057971,
+      "eval_recall": 0.7791666666666667,
+      "eval_runtime": 4.1183,
+      "eval_samples_per_second": 58.276,
+      "eval_steps_per_second": 0.971,
       "step": 36
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.9083333333333333,
-      "eval_f1_score": 0.8940810124891471,
-      "eval_loss": 0.24386708438396454,
-      "eval_precision": 0.9006296296296298,
-      "eval_recall": 0.9083333333333333,
-      "eval_runtime": 3.9449,
-      "eval_samples_per_second": 60.839,
-      "eval_steps_per_second": 1.014,
+      "eval_accuracy": 0.7583333333333333,
+      "eval_f1_score": 0.704514442129346,
+      "eval_loss": 0.5022971630096436,
+      "eval_precision": 0.7393393393393393,
+      "eval_recall": 0.7583333333333333,
+      "eval_runtime": 3.9137,
+      "eval_samples_per_second": 61.323,
+      "eval_steps_per_second": 1.022,
       "step": 40
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.8916666666666667,
-      "eval_f1_score": 0.8884444626028217,
-      "eval_loss": 0.25889885425567627,
-      "eval_precision": 0.8860502521300643,
-      "eval_recall": 0.8916666666666667,
-      "eval_runtime": 4.1005,
-      "eval_samples_per_second": 58.529,
-      "eval_steps_per_second": 0.975,
+      "eval_accuracy": 0.7708333333333334,
+      "eval_f1_score": 0.7434882865423094,
+      "eval_loss": 0.4910615086555481,
+      "eval_precision": 0.7506953593910115,
+      "eval_recall": 0.7708333333333334,
+      "eval_runtime": 3.9528,
+      "eval_samples_per_second": 60.716,
+      "eval_steps_per_second": 1.012,
       "step": 44
     },
     {
       "epoch": 11.25,
       "learning_rate": 1.388888888888889e-05,
-      "loss": 0.3017,
+      "loss": 0.4379,
       "step": 45
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.9083333333333333,
-      "eval_f1_score": 0.9024039297513475,
-      "eval_loss": 0.24283821880817413,
-      "eval_precision": 0.9004925532625392,
-      "eval_recall": 0.9083333333333333,
-      "eval_runtime": 4.2192,
-      "eval_samples_per_second": 56.883,
-      "eval_steps_per_second": 0.948,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_f1_score": 0.7513341298287536,
+      "eval_loss": 0.492078959941864,
+      "eval_precision": 0.7487322201607916,
+      "eval_recall": 0.7666666666666667,
+      "eval_runtime": 4.01,
+      "eval_samples_per_second": 59.85,
+      "eval_steps_per_second": 0.998,
       "step": 48
     },
     {
       "epoch": 13.0,
-      "eval_accuracy": 0.9,
-      "eval_f1_score": 0.8970256577872201,
-      "eval_loss": 0.2543129324913025,
-      "eval_precision": 0.894922622152669,
-      "eval_recall": 0.9,
-      "eval_runtime": 4.181,
-      "eval_samples_per_second": 57.402,
-      "eval_steps_per_second": 0.957,
+      "eval_accuracy": 0.7916666666666666,
+      "eval_f1_score": 0.7679595398368771,
+      "eval_loss": 0.4905916154384613,
+      "eval_precision": 0.7791738054445079,
+      "eval_recall": 0.7916666666666666,
+      "eval_runtime": 4.0131,
+      "eval_samples_per_second": 59.804,
+      "eval_steps_per_second": 0.997,
       "step": 52
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.8958333333333334,
-      "eval_f1_score": 0.8951038120948916,
-      "eval_loss": 0.2651337683200836,
-      "eval_precision": 0.8944246031746032,
-      "eval_recall": 0.8958333333333334,
-      "eval_runtime": 3.9408,
-      "eval_samples_per_second": 60.902,
-      "eval_steps_per_second": 1.015,
+      "eval_accuracy": 0.7875,
+      "eval_f1_score": 0.76446353296747,
+      "eval_loss": 0.49192750453948975,
+      "eval_precision": 0.7731010452961672,
+      "eval_recall": 0.7875,
+      "eval_runtime": 4.0774,
+      "eval_samples_per_second": 58.86,
+      "eval_steps_per_second": 0.981,
       "step": 56
     },
     {
       "epoch": 15.0,
       "learning_rate": 0.0,
-      "loss": 0.278,
+      "loss": 0.4003,
       "step": 60
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.8958333333333334,
-      "eval_f1_score": 0.8951038120948916,
-      "eval_loss": 0.2636790871620178,
-      "eval_precision": 0.8944246031746032,
-      "eval_recall": 0.8958333333333334,
-      "eval_runtime": 4.1679,
-      "eval_samples_per_second": 57.584,
-      "eval_steps_per_second": 0.96,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_f1_score": 0.7586779214303522,
+      "eval_loss": 0.49290910363197327,
+      "eval_precision": 0.7677707976394442,
+      "eval_recall": 0.7833333333333333,
+      "eval_runtime": 3.978,
+      "eval_samples_per_second": 60.332,
+      "eval_steps_per_second": 1.006,
       "step": 60
     },
     {
       "epoch": 15.0,
       "step": 60,
       "total_flos": 1.16237984421888e+18,
-      "train_loss": 0.3610760450363159,
-      "train_runtime": 974.8105,
-      "train_samples_per_second": 15.388,
-      "train_steps_per_second": 0.062
+      "train_loss": 0.475440772374471,
+      "train_runtime": 1200.7807,
+      "train_samples_per_second": 12.492,
+      "train_steps_per_second": 0.05
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.9033333333333333,
-      "eval_f1_score": 0.888913443830571,
-      "eval_loss": 0.24460919201374054,
-      "eval_precision": 0.892075919335706,
-      "eval_recall": 0.9033333333333333,
-      "eval_runtime": 4.8461,
-      "eval_samples_per_second": 61.906,
-      "eval_steps_per_second": 1.032,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_f1_score": 0.7558981348884555,
+      "eval_loss": 0.4771976172924042,
+      "eval_precision": 0.7701923076923076,
+      "eval_recall": 0.7833333333333333,
+      "eval_runtime": 93.6348,
+      "eval_samples_per_second": 3.204,
+      "eval_steps_per_second": 0.053,
       "step": 60
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e3f28fa49c01ce58eb2cbd98f07b83111f96e4736164481dabe0c27dccfd18c
+oid sha256:d7a4354514de5bcf1aa51ba982ed563501930d8c01edcbecc19581c7fc1d5d8d
 size 4027