End of training

README.md

@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0504
+- Loss: 0.0416
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 8.42,
+    "eval_loss": 0.041566625237464905,
+    "eval_runtime": 3.1481,
+    "eval_samples_per_second": 42.248,
+    "eval_steps_per_second": 0.953,
+    "total_flos": 3.274254821086986e+17,
+    "train_loss": 0.2961744487285614,
+    "train_runtime": 328.2191,
+    "train_samples_per_second": 36.378,
+    "train_steps_per_second": 0.122
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 8.42,
+    "eval_loss": 0.041566625237464905,
+    "eval_runtime": 3.1481,
+    "eval_samples_per_second": 42.248,
+    "eval_steps_per_second": 0.953
+}

runs/Apr23_14-39-00_a738cf30c8c2/events.out.tfevents.1713883484.a738cf30c8c2.34.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f4e3e308c2ffb56b98214950a22d6c695bac59f4d946cefb565dbcd3d86ddbe
+size 354

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.42,
+    "total_flos": 3.274254821086986e+17,
+    "train_loss": 0.2961744487285614,
+    "train_runtime": 328.2191,
+    "train_samples_per_second": 36.378,
+    "train_steps_per_second": 0.122
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.041566625237464905,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-Kontur-competition-1.3K/checkpoint-33",
+  "epoch": 8.421052631578947,
+  "eval_steps": 500,
+  "global_step": 40,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.84,
+      "eval_loss": 0.6306216716766357,
+      "eval_runtime": 5.0675,
+      "eval_samples_per_second": 26.246,
+      "eval_steps_per_second": 0.592,
+      "step": 4
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.41851410269737244,
+      "eval_runtime": 3.8343,
+      "eval_samples_per_second": 34.687,
+      "eval_steps_per_second": 0.782,
+      "step": 9
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 2.0625433921813965,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.6216,
+      "step": 10
+    },
+    {
+      "epoch": 2.95,
+      "eval_loss": 0.23511576652526855,
+      "eval_runtime": 3.8543,
+      "eval_samples_per_second": 34.507,
+      "eval_steps_per_second": 0.778,
+      "step": 14
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.12351791560649872,
+      "eval_runtime": 2.7637,
+      "eval_samples_per_second": 48.123,
+      "eval_steps_per_second": 1.085,
+      "step": 19
+    },
+    {
+      "epoch": 4.21,
+      "grad_norm": 2.8277649879455566,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.3128,
+      "step": 20
+    },
+    {
+      "epoch": 4.84,
+      "eval_loss": 0.07665743678808212,
+      "eval_runtime": 2.651,
+      "eval_samples_per_second": 50.171,
+      "eval_steps_per_second": 1.132,
+      "step": 23
+    },
+    {
+      "epoch": 5.89,
+      "eval_loss": 0.06760228425264359,
+      "eval_runtime": 2.7021,
+      "eval_samples_per_second": 49.22,
+      "eval_steps_per_second": 1.11,
+      "step": 28
+    },
+    {
+      "epoch": 6.32,
+      "grad_norm": 4.639766216278076,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.1492,
+      "step": 30
+    },
+    {
+      "epoch": 6.95,
+      "eval_loss": 0.041566625237464905,
+      "eval_runtime": 2.6918,
+      "eval_samples_per_second": 49.41,
+      "eval_steps_per_second": 1.115,
+      "step": 33
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.04927726835012436,
+      "eval_runtime": 2.6173,
+      "eval_samples_per_second": 50.815,
+      "eval_steps_per_second": 1.146,
+      "step": 38
+    },
+    {
+      "epoch": 8.42,
+      "grad_norm": 3.769456624984741,
+      "learning_rate": 0.0,
+      "loss": 0.1011,
+      "step": 40
+    },
+    {
+      "epoch": 8.42,
+      "eval_loss": 0.05041909217834473,
+      "eval_runtime": 3.2185,
+      "eval_samples_per_second": 41.323,
+      "eval_steps_per_second": 0.932,
+      "step": 40
+    },
+    {
+      "epoch": 8.42,
+      "step": 40,
+      "total_flos": 3.274254821086986e+17,
+      "train_loss": 0.2961744487285614,
+      "train_runtime": 328.2191,
+      "train_samples_per_second": 36.378,
+      "train_steps_per_second": 0.122
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 40,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 3.274254821086986e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}