End of training

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 datasets:
 - imagefolder

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 100.0,
-    "eval_accuracy": 0.5461538461538461,
-    "eval_loss": 2.859877347946167,
-    "eval_runtime": 2.5644,
-    "eval_samples_per_second": 50.694,
-    "eval_steps_per_second": 0.78,
-    "train_loss": 2.6316724504743303,
-    "train_runtime": 4344.3701,
-    "train_samples_per_second": 11.97,
-    "train_steps_per_second": 0.161
+    "epoch": 120.0,
+    "eval_accuracy": 0.5846153846153846,
+    "eval_loss": 2.5402629375457764,
+    "eval_runtime": 2.5184,
+    "eval_samples_per_second": 51.619,
+    "eval_steps_per_second": 0.794,
+    "train_loss": 0.3072653747740246,
+    "train_runtime": 892.0889,
+    "train_samples_per_second": 69.948,
+    "train_steps_per_second": 0.942
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "eval_accuracy": 0.5461538461538461,
-    "eval_loss": 2.859877347946167,
-    "eval_runtime": 2.5644,
-    "eval_samples_per_second": 50.694,
-    "eval_steps_per_second": 0.78
+    "epoch": 120.0,
+    "eval_accuracy": 0.5846153846153846,
+    "eval_loss": 2.5402629375457764,
+    "eval_runtime": 2.5184,
+    "eval_samples_per_second": 51.619,
+    "eval_steps_per_second": 0.794
 }

runs/Jan17_19-49-28_d06676088071/events.out.tfevents.1705521910.d06676088071.25430.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8e1b9986a9eded89b9fae2faf40d5fd9e7da1fb8a2921cc370881f18a4eb1f4
+size 411

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 100.0,
-    "train_loss": 2.6316724504743303,
-    "train_runtime": 4344.3701,
-    "train_samples_per_second": 11.97,
-    "train_steps_per_second": 0.161
+    "epoch": 120.0,
+    "train_loss": 0.3072653747740246,
+    "train_runtime": 892.0889,
+    "train_samples_per_second": 69.948,
+    "train_steps_per_second": 0.942
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.859877347946167,
-  "best_model_checkpoint": "./drive/MyDrive/repositories/torch_example_image-classification/outputs_letter3/checkpoint-644",
-  "epoch": 100.0,
+  "best_metric": 2.5402629375457764,
+  "best_model_checkpoint": "./drive/MyDrive/repositories/torch_example_image-classification/outputs_letter3/checkpoint-840",
+  "epoch": 120.0,
   "eval_steps": 500,
-  "global_step": 700,
+  "global_step": 840,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1336,14 +1336,287 @@
       "train_runtime": 4344.3701,
       "train_samples_per_second": 11.97,
       "train_steps_per_second": 0.161
+    },
+    {
+      "epoch": 101.0,
+      "eval_accuracy": 0.5153846153846153,
+      "eval_loss": 2.84305477142334,
+      "eval_runtime": 2.1617,
+      "eval_samples_per_second": 60.137,
+      "eval_steps_per_second": 0.925,
+      "step": 707
+    },
+    {
+      "epoch": 101.43,
+      "learning_rate": 1.9761904761904763e-05,
+      "loss": 2.067,
+      "step": 710
+    },
+    {
+      "epoch": 102.0,
+      "eval_accuracy": 0.5,
+      "eval_loss": 2.8063366413116455,
+      "eval_runtime": 2.1563,
+      "eval_samples_per_second": 60.288,
+      "eval_steps_per_second": 0.928,
+      "step": 714
+    },
+    {
+      "epoch": 102.86,
+      "learning_rate": 1.9523809523809524e-05,
+      "loss": 2.0234,
+      "step": 720
+    },
+    {
+      "epoch": 103.0,
+      "eval_accuracy": 0.5230769230769231,
+      "eval_loss": 2.7845773696899414,
+      "eval_runtime": 2.1676,
+      "eval_samples_per_second": 59.974,
+      "eval_steps_per_second": 0.923,
+      "step": 721
+    },
+    {
+      "epoch": 104.0,
+      "eval_accuracy": 0.5307692307692308,
+      "eval_loss": 2.762021541595459,
+      "eval_runtime": 2.2767,
+      "eval_samples_per_second": 57.1,
+      "eval_steps_per_second": 0.878,
+      "step": 728
+    },
+    {
+      "epoch": 104.29,
+      "learning_rate": 1.928571428571429e-05,
+      "loss": 1.9979,
+      "step": 730
+    },
+    {
+      "epoch": 105.0,
+      "eval_accuracy": 0.5538461538461539,
+      "eval_loss": 2.7421603202819824,
+      "eval_runtime": 2.2859,
+      "eval_samples_per_second": 56.871,
+      "eval_steps_per_second": 0.875,
+      "step": 735
+    },
+    {
+      "epoch": 105.71,
+      "learning_rate": 1.904761904761905e-05,
+      "loss": 1.9515,
+      "step": 740
+    },
+    {
+      "epoch": 106.0,
+      "eval_accuracy": 0.5615384615384615,
+      "eval_loss": 2.722872734069824,
+      "eval_runtime": 2.3487,
+      "eval_samples_per_second": 55.35,
+      "eval_steps_per_second": 0.852,
+      "step": 742
+    },
+    {
+      "epoch": 107.0,
+      "eval_accuracy": 0.5384615384615384,
+      "eval_loss": 2.7145817279815674,
+      "eval_runtime": 2.3976,
+      "eval_samples_per_second": 54.221,
+      "eval_steps_per_second": 0.834,
+      "step": 749
+    },
+    {
+      "epoch": 107.14,
+      "learning_rate": 1.880952380952381e-05,
+      "loss": 1.9095,
+      "step": 750
+    },
+    {
+      "epoch": 108.0,
+      "eval_accuracy": 0.5538461538461539,
+      "eval_loss": 2.6905264854431152,
+      "eval_runtime": 2.3887,
+      "eval_samples_per_second": 54.422,
+      "eval_steps_per_second": 0.837,
+      "step": 756
+    },
+    {
+      "epoch": 108.57,
+      "learning_rate": 1.8571428571428575e-05,
+      "loss": 1.8759,
+      "step": 760
+    },
+    {
+      "epoch": 109.0,
+      "eval_accuracy": 0.5538461538461539,
+      "eval_loss": 2.6840174198150635,
+      "eval_runtime": 2.369,
+      "eval_samples_per_second": 54.876,
+      "eval_steps_per_second": 0.844,
+      "step": 763
+    },
+    {
+      "epoch": 110.0,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 1.8561,
+      "step": 770
+    },
+    {
+      "epoch": 110.0,
+      "eval_accuracy": 0.5461538461538461,
+      "eval_loss": 2.6554973125457764,
+      "eval_runtime": 2.3964,
+      "eval_samples_per_second": 54.248,
+      "eval_steps_per_second": 0.835,
+      "step": 770
+    },
+    {
+      "epoch": 111.0,
+      "eval_accuracy": 0.5692307692307692,
+      "eval_loss": 2.6465163230895996,
+      "eval_runtime": 2.5418,
+      "eval_samples_per_second": 51.146,
+      "eval_steps_per_second": 0.787,
+      "step": 777
+    },
+    {
+      "epoch": 111.43,
+      "learning_rate": 1.8095238095238097e-05,
+      "loss": 1.8147,
+      "step": 780
+    },
+    {
+      "epoch": 112.0,
+      "eval_accuracy": 0.5384615384615384,
+      "eval_loss": 2.6321260929107666,
+      "eval_runtime": 2.6607,
+      "eval_samples_per_second": 48.859,
+      "eval_steps_per_second": 0.752,
+      "step": 784
+    },
+    {
+      "epoch": 112.86,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 1.799,
+      "step": 790
+    },
+    {
+      "epoch": 113.0,
+      "eval_accuracy": 0.5384615384615384,
+      "eval_loss": 2.6352975368499756,
+      "eval_runtime": 2.3449,
+      "eval_samples_per_second": 55.439,
+      "eval_steps_per_second": 0.853,
+      "step": 791
+    },
+    {
+      "epoch": 114.0,
+      "eval_accuracy": 0.5461538461538461,
+      "eval_loss": 2.6188342571258545,
+      "eval_runtime": 2.5585,
+      "eval_samples_per_second": 50.811,
+      "eval_steps_per_second": 0.782,
+      "step": 798
+    },
+    {
+      "epoch": 114.29,
+      "learning_rate": 1.761904761904762e-05,
+      "loss": 1.7536,
+      "step": 800
+    },
+    {
+      "epoch": 115.0,
+      "eval_accuracy": 0.5461538461538461,
+      "eval_loss": 2.6063857078552246,
+      "eval_runtime": 2.5441,
+      "eval_samples_per_second": 51.098,
+      "eval_steps_per_second": 0.786,
+      "step": 805
+    },
+    {
+      "epoch": 115.71,
+      "learning_rate": 1.7380952380952384e-05,
+      "loss": 1.7162,
+      "step": 810
+    },
+    {
+      "epoch": 116.0,
+      "eval_accuracy": 0.5538461538461539,
+      "eval_loss": 2.591280937194824,
+      "eval_runtime": 2.5836,
+      "eval_samples_per_second": 50.316,
+      "eval_steps_per_second": 0.774,
+      "step": 812
+    },
+    {
+      "epoch": 117.0,
+      "eval_accuracy": 0.5923076923076923,
+      "eval_loss": 2.5749051570892334,
+      "eval_runtime": 2.5441,
+      "eval_samples_per_second": 51.099,
+      "eval_steps_per_second": 0.786,
+      "step": 819
+    },
+    {
+      "epoch": 117.14,
+      "learning_rate": 1.7142857142857142e-05,
+      "loss": 1.7225,
+      "step": 820
+    },
+    {
+      "epoch": 118.0,
+      "eval_accuracy": 0.5846153846153846,
+      "eval_loss": 2.563140869140625,
+      "eval_runtime": 2.6117,
+      "eval_samples_per_second": 49.777,
+      "eval_steps_per_second": 0.766,
+      "step": 826
+    },
+    {
+      "epoch": 118.57,
+      "learning_rate": 1.6904761904761906e-05,
+      "loss": 1.6841,
+      "step": 830
+    },
+    {
+      "epoch": 119.0,
+      "eval_accuracy": 0.5923076923076923,
+      "eval_loss": 2.547883987426758,
+      "eval_runtime": 2.5689,
+      "eval_samples_per_second": 50.605,
+      "eval_steps_per_second": 0.779,
+      "step": 833
+    },
+    {
+      "epoch": 120.0,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 1.6388,
+      "step": 840
+    },
+    {
+      "epoch": 120.0,
+      "eval_accuracy": 0.5846153846153846,
+      "eval_loss": 2.5402629375457764,
+      "eval_runtime": 2.4843,
+      "eval_samples_per_second": 52.328,
+      "eval_steps_per_second": 0.805,
+      "step": 840
+    },
+    {
+      "epoch": 120.0,
+      "step": 840,
+      "total_flos": 4.837667104869581e+18,
+      "train_loss": 0.3072653747740246,
+      "train_runtime": 892.0889,
+      "train_samples_per_second": 69.948,
+      "train_steps_per_second": 0.942
     }
   ],
   "logging_steps": 10,
-  "max_steps": 700,
+  "max_steps": 840,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 100,
+  "num_train_epochs": 120,
   "save_steps": 500,
-  "total_flos": 4.031389254057984e+18,
+  "total_flos": 4.837667104869581e+18,
   "train_batch_size": 80,
   "trial_name": null,
   "trial_params": null