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/swin-base-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-base-patch4-window7-224-in22k) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1231
+- Loss: 0.0549
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 8.42,
+    "eval_loss": 0.05487216264009476,
+    "eval_runtime": 2.638,
+    "eval_samples_per_second": 50.417,
+    "eval_steps_per_second": 1.137,
+    "total_flos": 7.88469671840514e+17,
+    "train_loss": 0.18382581882178783,
+    "train_runtime": 367.9939,
+    "train_samples_per_second": 32.446,
+    "train_steps_per_second": 0.109
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 8.42,
+    "eval_loss": 0.05487216264009476,
+    "eval_runtime": 2.638,
+    "eval_samples_per_second": 50.417,
+    "eval_steps_per_second": 1.137
+}

runs/Apr23_14-07-19_2898e28fdb14/events.out.tfevents.1713881649.2898e28fdb14.34.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20a58969ed292ea7ff6bb419ad6cbdd992f61b19a1929f6cd4ad878803243d32
+size 354

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.42,
+    "total_flos": 7.88469671840514e+17,
+    "train_loss": 0.18382581882178783,
+    "train_runtime": 367.9939,
+    "train_samples_per_second": 32.446,
+    "train_steps_per_second": 0.109
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.05487216264009476,
+  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-Kontur-competition-1.4K/checkpoint-23",
+  "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": 0.5903199911117554,
+      "eval_runtime": 3.2368,
+      "eval_samples_per_second": 41.09,
+      "eval_steps_per_second": 0.927,
+      "step": 4
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.32031941413879395,
+      "eval_runtime": 2.5544,
+      "eval_samples_per_second": 52.067,
+      "eval_steps_per_second": 1.174,
+      "step": 9
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 12.904247283935547,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.5582,
+      "step": 10
+    },
+    {
+      "epoch": 2.95,
+      "eval_loss": 0.2519628703594208,
+      "eval_runtime": 2.7088,
+      "eval_samples_per_second": 49.1,
+      "eval_steps_per_second": 1.108,
+      "step": 14
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.09062427282333374,
+      "eval_runtime": 2.6817,
+      "eval_samples_per_second": 49.596,
+      "eval_steps_per_second": 1.119,
+      "step": 19
+    },
+    {
+      "epoch": 4.21,
+      "grad_norm": 2.254042148590088,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.1094,
+      "step": 20
+    },
+    {
+      "epoch": 4.84,
+      "eval_loss": 0.05487216264009476,
+      "eval_runtime": 2.5622,
+      "eval_samples_per_second": 51.908,
+      "eval_steps_per_second": 1.171,
+      "step": 23
+    },
+    {
+      "epoch": 5.89,
+      "eval_loss": 0.14387370645999908,
+      "eval_runtime": 2.8209,
+      "eval_samples_per_second": 47.149,
+      "eval_steps_per_second": 1.064,
+      "step": 28
+    },
+    {
+      "epoch": 6.32,
+      "grad_norm": 2.9556210041046143,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.0533,
+      "step": 30
+    },
+    {
+      "epoch": 6.95,
+      "eval_loss": 0.06752961128950119,
+      "eval_runtime": 2.5525,
+      "eval_samples_per_second": 52.106,
+      "eval_steps_per_second": 1.175,
+      "step": 33
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.12144926190376282,
+      "eval_runtime": 2.5532,
+      "eval_samples_per_second": 52.091,
+      "eval_steps_per_second": 1.175,
+      "step": 38
+    },
+    {
+      "epoch": 8.42,
+      "grad_norm": 0.5887470245361328,
+      "learning_rate": 0.0,
+      "loss": 0.0144,
+      "step": 40
+    },
+    {
+      "epoch": 8.42,
+      "eval_loss": 0.1230710819363594,
+      "eval_runtime": 2.6049,
+      "eval_samples_per_second": 51.058,
+      "eval_steps_per_second": 1.152,
+      "step": 40
+    },
+    {
+      "epoch": 8.42,
+      "step": 40,
+      "total_flos": 7.88469671840514e+17,
+      "train_loss": 0.18382581882178783,
+      "train_runtime": 367.9939,
+      "train_samples_per_second": 32.446,
+      "train_steps_per_second": 0.109
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 40,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 7.88469671840514e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}