End of training

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.95,
+    "eval_accuracy": 0.8868159203980099,
+    "eval_loss": 0.3560946583747864,
+    "eval_runtime": 9.9975,
+    "eval_samples_per_second": 80.42,
+    "eval_steps_per_second": 2.601,
+    "total_flos": 3.961714492511355e+17,
+    "train_loss": 0.8394510049819947,
+    "train_runtime": 2517.839,
+    "train_samples_per_second": 6.386,
+    "train_steps_per_second": 0.05
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.95,
+    "eval_accuracy": 0.8868159203980099,
+    "eval_loss": 0.3560946583747864,
+    "eval_runtime": 9.9975,
+    "eval_samples_per_second": 80.42,
+    "eval_steps_per_second": 2.601
+}

runs/Apr17_07-02-23_fae8031ca2fe/events.out.tfevents.1713339931.fae8031ca2fe.3274.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f136f347b63380785c411a1efd80941c001c936ec7add728fb17111566532b1
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.95,
+    "total_flos": 3.961714492511355e+17,
+    "train_loss": 0.8394510049819947,
+    "train_runtime": 2517.839,
+    "train_samples_per_second": 6.386,
+    "train_steps_per_second": 0.05
+}

trainer_state.json

@@ -0,0 +1,159 @@
+{
+  "best_metric": 0.8868159203980099,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-car0005-realimg/checkpoint-125",
+  "epoch": 4.9504950495049505,
+  "eval_steps": 500,
+  "global_step": 125,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.4,
+      "grad_norm": 4.037953853607178,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 2.061,
+      "step": 10
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 5.747158527374268,
+      "learning_rate": 4.6875e-05,
+      "loss": 1.7814,
+      "step": 20
+    },
+    {
+      "epoch": 0.99,
+      "eval_accuracy": 0.5696517412935324,
+      "eval_loss": 1.2809621095657349,
+      "eval_runtime": 468.0532,
+      "eval_samples_per_second": 1.718,
+      "eval_steps_per_second": 0.056,
+      "step": 25
+    },
+    {
+      "epoch": 1.19,
+      "grad_norm": 11.353270530700684,
+      "learning_rate": 4.2410714285714285e-05,
+      "loss": 1.2692,
+      "step": 30
+    },
+    {
+      "epoch": 1.58,
+      "grad_norm": 9.617542266845703,
+      "learning_rate": 3.794642857142857e-05,
+      "loss": 0.9368,
+      "step": 40
+    },
+    {
+      "epoch": 1.98,
+      "grad_norm": 20.671083450317383,
+      "learning_rate": 3.348214285714286e-05,
+      "loss": 0.8304,
+      "step": 50
+    },
+    {
+      "epoch": 1.98,
+      "eval_accuracy": 0.763681592039801,
+      "eval_loss": 0.6862999200820923,
+      "eval_runtime": 10.0686,
+      "eval_samples_per_second": 79.852,
+      "eval_steps_per_second": 2.582,
+      "step": 50
+    },
+    {
+      "epoch": 2.38,
+      "grad_norm": 9.588895797729492,
+      "learning_rate": 2.9017857142857146e-05,
+      "loss": 0.6374,
+      "step": 60
+    },
+    {
+      "epoch": 2.77,
+      "grad_norm": 5.951541900634766,
+      "learning_rate": 2.455357142857143e-05,
+      "loss": 0.5527,
+      "step": 70
+    },
+    {
+      "epoch": 2.97,
+      "eval_accuracy": 0.8432835820895522,
+      "eval_loss": 0.4411161541938782,
+      "eval_runtime": 10.0751,
+      "eval_samples_per_second": 79.801,
+      "eval_steps_per_second": 2.581,
+      "step": 75
+    },
+    {
+      "epoch": 3.17,
+      "grad_norm": 9.441178321838379,
+      "learning_rate": 2.0089285714285717e-05,
+      "loss": 0.5089,
+      "step": 80
+    },
+    {
+      "epoch": 3.56,
+      "grad_norm": 7.397172927856445,
+      "learning_rate": 1.5625e-05,
+      "loss": 0.4664,
+      "step": 90
+    },
+    {
+      "epoch": 3.96,
+      "grad_norm": 7.929540157318115,
+      "learning_rate": 1.1160714285714287e-05,
+      "loss": 0.4643,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8756218905472637,
+      "eval_loss": 0.3699795603752136,
+      "eval_runtime": 10.0734,
+      "eval_samples_per_second": 79.814,
+      "eval_steps_per_second": 2.581,
+      "step": 101
+    },
+    {
+      "epoch": 4.36,
+      "grad_norm": 7.214543342590332,
+      "learning_rate": 6.696428571428572e-06,
+      "loss": 0.3934,
+      "step": 110
+    },
+    {
+      "epoch": 4.75,
+      "grad_norm": 13.205345153808594,
+      "learning_rate": 2.2321428571428573e-06,
+      "loss": 0.3917,
+      "step": 120
+    },
+    {
+      "epoch": 4.95,
+      "eval_accuracy": 0.8868159203980099,
+      "eval_loss": 0.3560946583747864,
+      "eval_runtime": 10.0159,
+      "eval_samples_per_second": 80.272,
+      "eval_steps_per_second": 2.596,
+      "step": 125
+    },
+    {
+      "epoch": 4.95,
+      "step": 125,
+      "total_flos": 3.961714492511355e+17,
+      "train_loss": 0.8394510049819947,
+      "train_runtime": 2517.839,
+      "train_samples_per_second": 6.386,
+      "train_steps_per_second": 0.05
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 125,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 3.961714492511355e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}