🍻 cheers

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.38402849435806274,
+    "eval_runtime": 1.7644,
+    "eval_samples_per_second": 117.888,
+    "eval_steps_per_second": 7.368,
+    "total_flos": 3.22436019290112e+17,
+    "train_loss": 0.9762612545129025,
+    "train_runtime": 74.7912,
+    "train_samples_per_second": 55.622,
+    "train_steps_per_second": 1.765
+}

runs/Jan11_08-33-56_458fbaf9acf0/events.out.tfevents.1704962149.458fbaf9acf0.26.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ad22824670fb098090c73176f3327e87bfd61d6ac0941dc9148ca5453093715
+size 363

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.38402849435806274,
+    "eval_runtime": 1.7644,
+    "eval_samples_per_second": 117.888,
+    "eval_steps_per_second": 7.368
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "total_flos": 3.22436019290112e+17,
+    "train_loss": 0.9762612545129025,
+    "train_runtime": 74.7912,
+    "train_samples_per_second": 55.622,
+    "train_steps_per_second": 1.765
+}

trainer_state.json

@@ -0,0 +1,112 @@
+{
+  "best_metric": 0.38402849435806274,
+  "best_model_checkpoint": "./vit-base-uppercase-english-characters/checkpoint-100",
+  "epoch": 4.0,
+  "global_step": 132,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00018484848484848484,
+      "loss": 3.0027,
+      "step": 10
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00016969696969696972,
+      "loss": 2.2789,
+      "step": 20
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00015454545454545454,
+      "loss": 1.6762,
+      "step": 30
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0001393939393939394,
+      "loss": 1.2222,
+      "step": 40
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00012424242424242425,
+      "loss": 0.9241,
+      "step": 50
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00010909090909090909,
+      "loss": 0.7407,
+      "step": 60
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 9.393939393939395e-05,
+      "loss": 0.5808,
+      "step": 70
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 7.878787878787879e-05,
+      "loss": 0.4945,
+      "step": 80
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 0.4364,
+      "step": 90
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 4.848484848484849e-05,
+      "loss": 0.3936,
+      "step": 100
+    },
+    {
+      "epoch": 3.03,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.38402849435806274,
+      "eval_runtime": 1.6842,
+      "eval_samples_per_second": 123.504,
+      "eval_steps_per_second": 7.719,
+      "step": 100
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.3722,
+      "step": 110
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 0.3534,
+      "step": 120
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 3.0303030303030305e-06,
+      "loss": 0.3421,
+      "step": 130
+    },
+    {
+      "epoch": 4.0,
+      "step": 132,
+      "total_flos": 3.22436019290112e+17,
+      "train_loss": 0.9762612545129025,
+      "train_runtime": 74.7912,
+      "train_samples_per_second": 55.622,
+      "train_steps_per_second": 1.765
+    }
+  ],
+  "max_steps": 132,
+  "num_train_epochs": 4,
+  "total_flos": 3.22436019290112e+17,
+  "trial_name": null,
+  "trial_params": null
+}