End of training

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 10.0,
-    "eval_f1": 0.8607594936708861,
-    "eval_loss": 0.26272934675216675,
-    "eval_runtime": 10.1059,
-    "eval_samples_per_second": 7.817,
+    "epoch": 30.0,
+    "eval_f1": 0.9620253164556962,
+    "eval_loss": 0.12829144299030304,
+    "eval_runtime": 10.0743,
+    "eval_samples_per_second": 7.842,
     "eval_steps_per_second": 0.099,
-    "train_loss": 0.3480381011962891,
-    "train_runtime": 655.7401,
-    "train_samples_per_second": 6.801,
-    "train_steps_per_second": 0.107
+    "train_loss": 0.20286732230867657,
+    "train_runtime": 1939.2485,
+    "train_samples_per_second": 6.9,
+    "train_steps_per_second": 0.108
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_f1": 0.8607594936708861,
-    "eval_loss": 0.26272934675216675,
-    "eval_runtime": 10.1059,
-    "eval_samples_per_second": 7.817,
+    "epoch": 30.0,
+    "eval_f1": 0.9620253164556962,
+    "eval_loss": 0.12829144299030304,
+    "eval_runtime": 10.0743,
+    "eval_samples_per_second": 7.842,
     "eval_steps_per_second": 0.099
 }

runs/Dec06_19-59-00_49793c51f922/events.out.tfevents.1670359112.49793c51f922.371.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e397a2f826749209e647af9193c29c05a4fa40d7fa884bf7a7ee4a24d3164913
+size 357

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.3480381011962891,
-    "train_runtime": 655.7401,
-    "train_samples_per_second": 6.801,
-    "train_steps_per_second": 0.107
+    "epoch": 30.0,
+    "train_loss": 0.20286732230867657,
+    "train_runtime": 1939.2485,
+    "train_samples_per_second": 6.9,
+    "train_steps_per_second": 0.108
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.26272934675216675,
-  "best_model_checkpoint": "/leicester_binary_convnext_small6/checkpoint-70",
-  "epoch": 10.0,
-  "global_step": 70,
+  "best_metric": 0.12829144299030304,
+  "best_model_checkpoint": "/leicester_binary_convnext_small/checkpoint-182",
+  "epoch": 30.0,
+  "global_step": 210,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,148 +10,412 @@
     {
       "epoch": 1.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.5187062621116638,
-      "eval_runtime": 10.04,
-      "eval_samples_per_second": 7.869,
+      "eval_loss": 0.5143362283706665,
+      "eval_runtime": 10.0208,
+      "eval_samples_per_second": 7.884,
       "eval_steps_per_second": 0.1,
       "step": 7
     },
     {
       "epoch": 1.43,
-      "learning_rate": 1.7142857142857142e-05,
-      "loss": 0.5904,
+      "learning_rate": 1.904761904761905e-05,
+      "loss": 0.5872,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.4272507429122925,
-      "eval_runtime": 10.0147,
-      "eval_samples_per_second": 7.888,
-      "eval_steps_per_second": 0.1,
+      "eval_loss": 0.42147475481033325,
+      "eval_runtime": 10.225,
+      "eval_samples_per_second": 7.726,
+      "eval_steps_per_second": 0.098,
       "step": 14
     },
     {
       "epoch": 2.86,
-      "learning_rate": 1.4285714285714287e-05,
-      "loss": 0.3981,
+      "learning_rate": 1.8095238095238097e-05,
+      "loss": 0.3903,
       "step": 20
     },
     {
       "epoch": 3.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.411451131105423,
-      "eval_runtime": 10.0801,
-      "eval_samples_per_second": 7.837,
-      "eval_steps_per_second": 0.099,
+      "eval_loss": 0.4127243757247925,
+      "eval_runtime": 9.9737,
+      "eval_samples_per_second": 7.921,
+      "eval_steps_per_second": 0.1,
       "step": 21
     },
     {
       "epoch": 4.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.40291550755500793,
-      "eval_runtime": 10.1912,
-      "eval_samples_per_second": 7.752,
-      "eval_steps_per_second": 0.098,
+      "eval_loss": 0.36054694652557373,
+      "eval_runtime": 9.9338,
+      "eval_samples_per_second": 7.953,
+      "eval_steps_per_second": 0.101,
       "step": 28
     },
     {
       "epoch": 4.29,
-      "learning_rate": 1.1428571428571429e-05,
-      "loss": 0.3285,
+      "learning_rate": 1.7142857142857142e-05,
+      "loss": 0.3163,
       "step": 30
     },
     {
       "epoch": 5.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.3401729464530945,
-      "eval_runtime": 10.0131,
-      "eval_samples_per_second": 7.89,
-      "eval_steps_per_second": 0.1,
+      "eval_loss": 0.3152060806751251,
+      "eval_runtime": 10.1128,
+      "eval_samples_per_second": 7.812,
+      "eval_steps_per_second": 0.099,
       "step": 35
     },
     {
       "epoch": 5.71,
-      "learning_rate": 8.571428571428571e-06,
-      "loss": 0.308,
+      "learning_rate": 1.6190476190476193e-05,
+      "loss": 0.2942,
       "step": 40
     },
     {
       "epoch": 6.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.31382453441619873,
-      "eval_runtime": 10.0365,
-      "eval_samples_per_second": 7.871,
+      "eval_loss": 0.2942211925983429,
+      "eval_runtime": 10.0302,
+      "eval_samples_per_second": 7.876,
       "eval_steps_per_second": 0.1,
       "step": 42
     },
     {
       "epoch": 7.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.29120802879333496,
-      "eval_runtime": 10.1307,
-      "eval_samples_per_second": 7.798,
-      "eval_steps_per_second": 0.099,
+      "eval_loss": 0.2668735980987549,
+      "eval_runtime": 9.9387,
+      "eval_samples_per_second": 7.949,
+      "eval_steps_per_second": 0.101,
       "step": 49
     },
     {
       "epoch": 7.14,
-      "learning_rate": 5.7142857142857145e-06,
-      "loss": 0.2952,
+      "learning_rate": 1.523809523809524e-05,
+      "loss": 0.2755,
       "step": 50
     },
     {
       "epoch": 8.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.2751685678958893,
-      "eval_runtime": 9.9617,
-      "eval_samples_per_second": 7.93,
-      "eval_steps_per_second": 0.1,
+      "eval_loss": 0.23157073557376862,
+      "eval_runtime": 10.0692,
+      "eval_samples_per_second": 7.846,
+      "eval_steps_per_second": 0.099,
       "step": 56
     },
     {
       "epoch": 8.57,
-      "learning_rate": 2.8571428571428573e-06,
-      "loss": 0.2593,
+      "learning_rate": 1.4285714285714287e-05,
+      "loss": 0.2281,
       "step": 60
     },
     {
       "epoch": 9.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.2657069265842438,
-      "eval_runtime": 10.0549,
-      "eval_samples_per_second": 7.857,
+      "eval_loss": 0.2104099839925766,
+      "eval_runtime": 10.0928,
+      "eval_samples_per_second": 7.827,
       "eval_steps_per_second": 0.099,
       "step": 63
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.0,
-      "loss": 0.2568,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.2076,
       "step": 70
     },
     {
       "epoch": 10.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.26272934675216675,
-      "eval_runtime": 10.516,
-      "eval_samples_per_second": 7.512,
-      "eval_steps_per_second": 0.095,
+      "eval_loss": 0.1937984824180603,
+      "eval_runtime": 9.9756,
+      "eval_samples_per_second": 7.919,
+      "eval_steps_per_second": 0.1,
       "step": 70
     },
     {
-      "epoch": 10.0,
-      "step": 70,
-      "total_flos": 1.9921607202889728e+17,
-      "train_loss": 0.3480381011962891,
-      "train_runtime": 655.7401,
-      "train_samples_per_second": 6.801,
-      "train_steps_per_second": 0.107
+      "epoch": 11.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.1803407520055771,
+      "eval_runtime": 10.0131,
+      "eval_samples_per_second": 7.89,
+      "eval_steps_per_second": 0.1,
+      "step": 77
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 1.2380952380952383e-05,
+      "loss": 0.1832,
+      "step": 80
+    },
+    {
+      "epoch": 12.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.17044714093208313,
+      "eval_runtime": 10.0514,
+      "eval_samples_per_second": 7.86,
+      "eval_steps_per_second": 0.099,
+      "step": 84
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.1523809523809524e-05,
+      "loss": 0.1758,
+      "step": 90
+    },
+    {
+      "epoch": 13.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.16500937938690186,
+      "eval_runtime": 9.8999,
+      "eval_samples_per_second": 7.98,
+      "eval_steps_per_second": 0.101,
+      "step": 91
+    },
+    {
+      "epoch": 14.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.17136065661907196,
+      "eval_runtime": 10.3084,
+      "eval_samples_per_second": 7.664,
+      "eval_steps_per_second": 0.097,
+      "step": 98
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 1.0571428571428572e-05,
+      "loss": 0.167,
+      "step": 100
+    },
+    {
+      "epoch": 15.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.1574719250202179,
+      "eval_runtime": 10.3458,
+      "eval_samples_per_second": 7.636,
+      "eval_steps_per_second": 0.097,
+      "step": 105
+    },
+    {
+      "epoch": 15.71,
+      "learning_rate": 9.61904761904762e-06,
+      "loss": 0.1519,
+      "step": 110
+    },
+    {
+      "epoch": 16.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.154911071062088,
+      "eval_runtime": 9.9419,
+      "eval_samples_per_second": 7.946,
+      "eval_steps_per_second": 0.101,
+      "step": 112
+    },
+    {
+      "epoch": 17.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.17046359181404114,
+      "eval_runtime": 10.1139,
+      "eval_samples_per_second": 7.811,
+      "eval_steps_per_second": 0.099,
+      "step": 119
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.1422,
+      "step": 120
+    },
+    {
+      "epoch": 18.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.14778320491313934,
+      "eval_runtime": 10.4334,
+      "eval_samples_per_second": 7.572,
+      "eval_steps_per_second": 0.096,
+      "step": 126
+    },
+    {
+      "epoch": 18.57,
+      "learning_rate": 7.714285714285716e-06,
+      "loss": 0.1444,
+      "step": 130
+    },
+    {
+      "epoch": 19.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.14368951320648193,
+      "eval_runtime": 9.9263,
+      "eval_samples_per_second": 7.959,
+      "eval_steps_per_second": 0.101,
+      "step": 133
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 6.761904761904763e-06,
+      "loss": 0.1396,
+      "step": 140
+    },
+    {
+      "epoch": 20.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.13980019092559814,
+      "eval_runtime": 10.0444,
+      "eval_samples_per_second": 7.865,
+      "eval_steps_per_second": 0.1,
+      "step": 140
+    },
+    {
+      "epoch": 21.0,
+      "eval_f1": 0.8607594936708861,
+      "eval_loss": 0.13507133722305298,
+      "eval_runtime": 9.9798,
+      "eval_samples_per_second": 7.916,
+      "eval_steps_per_second": 0.1,
+      "step": 147
+    },
+    {
+      "epoch": 21.43,
+      "learning_rate": 5.8095238095238106e-06,
+      "loss": 0.1293,
+      "step": 150
+    },
+    {
+      "epoch": 22.0,
+      "eval_f1": 0.8987341772151899,
+      "eval_loss": 0.1370120793581009,
+      "eval_runtime": 9.9423,
+      "eval_samples_per_second": 7.946,
+      "eval_steps_per_second": 0.101,
+      "step": 154
+    },
+    {
+      "epoch": 22.86,
+      "learning_rate": 4.857142857142858e-06,
+      "loss": 0.1361,
+      "step": 160
+    },
+    {
+      "epoch": 23.0,
+      "eval_f1": 0.8987341772151899,
+      "eval_loss": 0.13351036608219147,
+      "eval_runtime": 9.9968,
+      "eval_samples_per_second": 7.903,
+      "eval_steps_per_second": 0.1,
+      "step": 161
+    },
+    {
+      "epoch": 24.0,
+      "eval_f1": 0.9367088607594937,
+      "eval_loss": 0.13105367124080658,
+      "eval_runtime": 10.0843,
+      "eval_samples_per_second": 7.834,
+      "eval_steps_per_second": 0.099,
+      "step": 168
+    },
+    {
+      "epoch": 24.29,
+      "learning_rate": 3.9047619047619055e-06,
+      "loss": 0.1246,
+      "step": 170
+    },
+    {
+      "epoch": 25.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.12886276841163635,
+      "eval_runtime": 9.9441,
+      "eval_samples_per_second": 7.944,
+      "eval_steps_per_second": 0.101,
+      "step": 175
+    },
+    {
+      "epoch": 25.71,
+      "learning_rate": 2.9523809523809525e-06,
+      "loss": 0.1211,
+      "step": 180
+    },
+    {
+      "epoch": 26.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.12829144299030304,
+      "eval_runtime": 10.0697,
+      "eval_samples_per_second": 7.845,
+      "eval_steps_per_second": 0.099,
+      "step": 182
+    },
+    {
+      "epoch": 27.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.12941910326480865,
+      "eval_runtime": 10.0409,
+      "eval_samples_per_second": 7.868,
+      "eval_steps_per_second": 0.1,
+      "step": 189
+    },
+    {
+      "epoch": 27.14,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.1182,
+      "step": 190
+    },
+    {
+      "epoch": 28.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.13063742220401764,
+      "eval_runtime": 9.9814,
+      "eval_samples_per_second": 7.915,
+      "eval_steps_per_second": 0.1,
+      "step": 196
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 1.0476190476190478e-06,
+      "loss": 0.1172,
+      "step": 200
+    },
+    {
+      "epoch": 29.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.13123980164527893,
+      "eval_runtime": 9.9869,
+      "eval_samples_per_second": 7.91,
+      "eval_steps_per_second": 0.1,
+      "step": 203
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 9.523809523809525e-08,
+      "loss": 0.1102,
+      "step": 210
+    },
+    {
+      "epoch": 30.0,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.1317748874425888,
+      "eval_runtime": 9.9991,
+      "eval_samples_per_second": 7.901,
+      "eval_steps_per_second": 0.1,
+      "step": 210
+    },
+    {
+      "epoch": 30.0,
+      "step": 210,
+      "total_flos": 5.976482160866918e+17,
+      "train_loss": 0.20286732230867657,
+      "train_runtime": 1939.2485,
+      "train_samples_per_second": 6.9,
+      "train_steps_per_second": 0.108
     }
   ],
-  "max_steps": 70,
-  "num_train_epochs": 10,
-  "total_flos": 1.9921607202889728e+17,
+  "max_steps": 210,
+  "num_train_epochs": 30,
+  "total_flos": 5.976482160866918e+17,
   "trial_name": null,
   "trial_params": null
 }