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: 0.6819
+- Loss: 0.6773
 - Accuracy: 0.7833
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 9.49,
+    "eval_accuracy": 0.7833333333333333,
+    "eval_loss": 0.6773229837417603,
+    "eval_runtime": 1.0904,
+    "eval_samples_per_second": 55.026,
+    "eval_steps_per_second": 1.834,
+    "train_loss": 0.6781736867768424,
+    "train_runtime": 289.6929,
+    "train_samples_per_second": 64.827,
+    "train_steps_per_second": 0.483
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.49,
+    "eval_accuracy": 0.7833333333333333,
+    "eval_loss": 0.6773229837417603,
+    "eval_runtime": 1.0904,
+    "eval_samples_per_second": 55.026,
+    "eval_steps_per_second": 1.834
+}

runs/May24_08-12-46_DESKTOP-SKBE9FB/events.out.tfevents.1716560337.DESKTOP-SKBE9FB.10952.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faebe14ce04854f86ebd49e953bd2a9550cb28ef30f6fb7fef905d3f6c2b6006
+size 411

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 9.49,
+    "train_loss": 0.6781736867768424,
+    "train_runtime": 289.6929,
+    "train_samples_per_second": 64.827,
+    "train_steps_per_second": 0.483
+}

trainer_state.json

@@ -0,0 +1,204 @@
+{
+  "best_metric": 0.7833333333333333,
+  "best_model_checkpoint": "vit-base-patch16-224-RU5-10-8\\checkpoint-88",
+  "epoch": 9.491525423728813,
+  "eval_steps": 500,
+  "global_step": 140,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.68,
+      "learning_rate": 5.37593984962406e-05,
+      "loss": 1.3605,
+      "step": 10
+    },
+    {
+      "epoch": 0.95,
+      "eval_accuracy": 0.5166666666666667,
+      "eval_loss": 1.2370423078536987,
+      "eval_runtime": 1.0132,
+      "eval_samples_per_second": 59.22,
+      "eval_steps_per_second": 1.974,
+      "step": 14
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4.9624060150375936e-05,
+      "loss": 1.2314,
+      "step": 20
+    },
+    {
+      "epoch": 1.97,
+      "eval_accuracy": 0.6833333333333333,
+      "eval_loss": 1.0510814189910889,
+      "eval_runtime": 1.0312,
+      "eval_samples_per_second": 58.187,
+      "eval_steps_per_second": 1.94,
+      "step": 29
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.548872180451128e-05,
+      "loss": 1.1186,
+      "step": 30
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.135338345864661e-05,
+      "loss": 0.968,
+      "step": 40
+    },
+    {
+      "epoch": 2.98,
+      "eval_accuracy": 0.65,
+      "eval_loss": 0.891896665096283,
+      "eval_runtime": 1.03,
+      "eval_samples_per_second": 58.251,
+      "eval_steps_per_second": 1.942,
+      "step": 44
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 3.7218045112781954e-05,
+      "loss": 0.8135,
+      "step": 50
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_loss": 0.7701808214187622,
+      "eval_runtime": 1.0529,
+      "eval_samples_per_second": 56.985,
+      "eval_steps_per_second": 1.899,
+      "step": 59
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.3082706766917295e-05,
+      "loss": 0.708,
+      "step": 60
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 2.894736842105263e-05,
+      "loss": 0.616,
+      "step": 70
+    },
+    {
+      "epoch": 4.95,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.7533265352249146,
+      "eval_runtime": 1.0123,
+      "eval_samples_per_second": 59.269,
+      "eval_steps_per_second": 1.976,
+      "step": 73
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 2.4812030075187968e-05,
+      "loss": 0.5167,
+      "step": 80
+    },
+    {
+      "epoch": 5.97,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6773229837417603,
+      "eval_runtime": 1.0296,
+      "eval_samples_per_second": 58.276,
+      "eval_steps_per_second": 1.943,
+      "step": 88
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 2.0676691729323306e-05,
+      "loss": 0.4584,
+      "step": 90
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 1.6541353383458648e-05,
+      "loss": 0.4063,
+      "step": 100
+    },
+    {
+      "epoch": 6.98,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.6973603963851929,
+      "eval_runtime": 1.0434,
+      "eval_samples_per_second": 57.507,
+      "eval_steps_per_second": 1.917,
+      "step": 103
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.2406015037593984e-05,
+      "loss": 0.3401,
+      "step": 110
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.7437816262245178,
+      "eval_runtime": 1.0454,
+      "eval_samples_per_second": 57.397,
+      "eval_steps_per_second": 1.913,
+      "step": 118
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 8.270676691729324e-06,
+      "loss": 0.3408,
+      "step": 120
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 4.135338345864662e-06,
+      "loss": 0.3007,
+      "step": 130
+    },
+    {
+      "epoch": 8.95,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6645744442939758,
+      "eval_runtime": 1.0356,
+      "eval_samples_per_second": 57.935,
+      "eval_steps_per_second": 1.931,
+      "step": 132
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 0.0,
+      "loss": 0.3154,
+      "step": 140
+    },
+    {
+      "epoch": 9.49,
+      "eval_accuracy": 0.7833333333333333,
+      "eval_loss": 0.6818878054618835,
+      "eval_runtime": 1.0224,
+      "eval_samples_per_second": 58.687,
+      "eval_steps_per_second": 1.956,
+      "step": 140
+    },
+    {
+      "epoch": 9.49,
+      "step": 140,
+      "total_flos": 1.381706941988782e+18,
+      "train_loss": 0.6781736867768424,
+      "train_runtime": 289.6929,
+      "train_samples_per_second": 64.827,
+      "train_steps_per_second": 0.483
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 140,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.381706941988782e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}