End of training

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6589
+- Loss: 0.6241
 - Accuracy: 0.7833
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 9.87,
+    "eval_accuracy": 0.7833333333333333,
+    "eval_loss": 0.624082624912262,
+    "eval_runtime": 1.3148,
+    "eval_samples_per_second": 45.634,
+    "eval_steps_per_second": 1.521,
+    "train_loss": 0.5378790767569291,
+    "train_runtime": 416.9022,
+    "train_samples_per_second": 58.767,
+    "train_steps_per_second": 0.456
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.87,
+    "eval_accuracy": 0.7833333333333333,
+    "eval_loss": 0.624082624912262,
+    "eval_runtime": 1.3148,
+    "eval_samples_per_second": 45.634,
+    "eval_steps_per_second": 1.521
+}

runs/May23_18-24-21_DESKTOP-SKBE9FB/events.out.tfevents.1716510810.DESKTOP-SKBE9FB.16448.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66ea7649cd2911c7b19b0eeb6dd9a188bb2c15d8e2a4b5551befe68a3baff326
+size 411

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 9.87,
+    "train_loss": 0.5378790767569291,
+    "train_runtime": 416.9022,
+    "train_samples_per_second": 58.767,
+    "train_steps_per_second": 0.456
+}

trainer_state.json

@@ -0,0 +1,234 @@
+{
+  "best_metric": 0.7833333333333333,
+  "best_model_checkpoint": "vit-base-patch16-224-RU3-10\\checkpoint-115",
+  "epoch": 9.87012987012987,
+  "eval_steps": 500,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.5e-05,
+      "loss": 1.3698,
+      "step": 10
+    },
+    {
+      "epoch": 0.99,
+      "eval_accuracy": 0.65,
+      "eval_loss": 1.1844545602798462,
+      "eval_runtime": 1.2572,
+      "eval_samples_per_second": 47.725,
+      "eval_steps_per_second": 1.591,
+      "step": 19
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 5.1944444444444446e-05,
+      "loss": 1.2391,
+      "step": 20
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.888888888888889e-05,
+      "loss": 1.1232,
+      "step": 30
+    },
+    {
+      "epoch": 1.97,
+      "eval_accuracy": 0.65,
+      "eval_loss": 0.9392767548561096,
+      "eval_runtime": 1.0792,
+      "eval_samples_per_second": 55.596,
+      "eval_steps_per_second": 1.853,
+      "step": 38
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.5833333333333334e-05,
+      "loss": 0.9316,
+      "step": 40
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.277777777777778e-05,
+      "loss": 0.8168,
+      "step": 50
+    },
+    {
+      "epoch": 2.96,
+      "eval_accuracy": 0.6333333333333333,
+      "eval_loss": 0.9117090702056885,
+      "eval_runtime": 1.1003,
+      "eval_samples_per_second": 54.533,
+      "eval_steps_per_second": 1.818,
+      "step": 57
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 3.972222222222222e-05,
+      "loss": 0.7242,
+      "step": 60
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.6666666666666666e-05,
+      "loss": 0.5992,
+      "step": 70
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_loss": 0.8329970240592957,
+      "eval_runtime": 1.1977,
+      "eval_samples_per_second": 50.098,
+      "eval_steps_per_second": 1.67,
+      "step": 77
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.3611111111111116e-05,
+      "loss": 0.5153,
+      "step": 80
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.4258,
+      "step": 90
+    },
+    {
+      "epoch": 4.99,
+      "eval_accuracy": 0.7,
+      "eval_loss": 0.7470608353614807,
+      "eval_runtime": 1.0986,
+      "eval_samples_per_second": 54.614,
+      "eval_steps_per_second": 1.82,
+      "step": 96
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 2.75e-05,
+      "loss": 0.3506,
+      "step": 100
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 2.4444444444444445e-05,
+      "loss": 0.3283,
+      "step": 110
+    },
+    {
+      "epoch": 5.97,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.624082624912262,
+      "eval_runtime": 1.1064,
+      "eval_samples_per_second": 54.229,
+      "eval_steps_per_second": 1.808,
+      "step": 115
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 2.138888888888889e-05,
+      "loss": 0.2722,
+      "step": 120
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.2543,
+      "step": 130
+    },
+    {
+      "epoch": 6.96,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.5915836691856384,
+      "eval_runtime": 1.0693,
+      "eval_samples_per_second": 56.109,
+      "eval_steps_per_second": 1.87,
+      "step": 134
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 1.527777777777778e-05,
+      "loss": 0.2577,
+      "step": 140
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 1.2222222222222222e-05,
+      "loss": 0.2345,
+      "step": 150
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6783313155174255,
+      "eval_runtime": 1.1474,
+      "eval_samples_per_second": 52.293,
+      "eval_steps_per_second": 1.743,
+      "step": 154
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 9.166666666666666e-06,
+      "loss": 0.2046,
+      "step": 160
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 6.111111111111111e-06,
+      "loss": 0.2027,
+      "step": 170
+    },
+    {
+      "epoch": 8.99,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6577182412147522,
+      "eval_runtime": 1.089,
+      "eval_samples_per_second": 55.098,
+      "eval_steps_per_second": 1.837,
+      "step": 173
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 3.0555555555555556e-06,
+      "loss": 0.1964,
+      "step": 180
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 0.0,
+      "loss": 0.1733,
+      "step": 190
+    },
+    {
+      "epoch": 9.87,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6588727235794067,
+      "eval_runtime": 1.1149,
+      "eval_samples_per_second": 53.819,
+      "eval_steps_per_second": 1.794,
+      "step": 190
+    },
+    {
+      "epoch": 9.87,
+      "step": 190,
+      "total_flos": 1.8748748039526973e+18,
+      "train_loss": 0.5378790767569291,
+      "train_runtime": 416.9022,
+      "train_samples_per_second": 58.767,
+      "train_steps_per_second": 0.456
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 190,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.8748748039526973e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}