End of training

README.md

@@ -18,8 +18,8 @@ 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 an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0548
-- Accuracy: 0.6154
+- Loss: 1.0777
+- Accuracy: 0.6374
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 21.0,
+    "eval_accuracy": 0.6373626373626373,
+    "eval_loss": 1.0777134895324707,
+    "eval_runtime": 1.4966,
+    "eval_samples_per_second": 60.803,
+    "eval_steps_per_second": 2.004,
+    "total_flos": 2.7877744402602394e+17,
+    "train_loss": 7.695562998453776,
+    "train_runtime": 380.562,
+    "train_samples_per_second": 45.028,
+    "train_steps_per_second": 0.11
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 21.0,
+    "eval_accuracy": 0.6373626373626373,
+    "eval_loss": 1.0777134895324707,
+    "eval_runtime": 1.4966,
+    "eval_samples_per_second": 60.803,
+    "eval_steps_per_second": 2.004
+}

runs/Feb06_01-04-38_2066b31ba251/events.out.tfevents.1738804302.2066b31ba251.21448.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58307e5480903378a4adf19426c639d1ff70edd0b0d1e00a17ed26d55be299b3
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 21.0,
+    "total_flos": 2.7877744402602394e+17,
+    "train_loss": 7.695562998453776,
+    "train_runtime": 380.562,
+    "train_samples_per_second": 45.028,
+    "train_steps_per_second": 0.11
+}

trainer_state.json

@@ -0,0 +1,252 @@
+{
+  "best_metric": 0.6373626373626373,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV43/checkpoint-30",
+  "epoch": 21.0,
+  "eval_steps": 500,
+  "global_step": 42,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.13186813186813187,
+      "eval_loss": 1.5998319387435913,
+      "eval_runtime": 1.5311,
+      "eval_samples_per_second": 59.434,
+      "eval_steps_per_second": 1.959,
+      "step": 2
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4175824175824176,
+      "eval_loss": 1.4959794282913208,
+      "eval_runtime": 1.8755,
+      "eval_samples_per_second": 48.521,
+      "eval_steps_per_second": 1.6,
+      "step": 4
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.43956043956043955,
+      "eval_loss": 1.365879774093628,
+      "eval_runtime": 1.4203,
+      "eval_samples_per_second": 64.071,
+      "eval_steps_per_second": 2.112,
+      "step": 6
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.42857142857142855,
+      "eval_loss": 1.3074045181274414,
+      "eval_runtime": 1.4332,
+      "eval_samples_per_second": 63.495,
+      "eval_steps_per_second": 2.093,
+      "step": 8
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.42857142857142855,
+      "eval_loss": 1.2849838733673096,
+      "eval_runtime": 1.3814,
+      "eval_samples_per_second": 65.875,
+      "eval_steps_per_second": 2.172,
+      "step": 10
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 17.214662551879883,
+      "learning_rate": 1.8285096492438424e-05,
+      "loss": 9.4768,
+      "step": 12
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.43956043956043955,
+      "eval_loss": 1.259225606918335,
+      "eval_runtime": 1.4313,
+      "eval_samples_per_second": 63.578,
+      "eval_steps_per_second": 2.096,
+      "step": 12
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.45054945054945056,
+      "eval_loss": 1.2445858716964722,
+      "eval_runtime": 1.5394,
+      "eval_samples_per_second": 59.113,
+      "eval_steps_per_second": 1.949,
+      "step": 14
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5494505494505495,
+      "eval_loss": 1.2326079607009888,
+      "eval_runtime": 1.9114,
+      "eval_samples_per_second": 47.61,
+      "eval_steps_per_second": 1.57,
+      "step": 16
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5714285714285714,
+      "eval_loss": 1.208442211151123,
+      "eval_runtime": 1.8328,
+      "eval_samples_per_second": 49.652,
+      "eval_steps_per_second": 1.637,
+      "step": 18
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5604395604395604,
+      "eval_loss": 1.1779202222824097,
+      "eval_runtime": 1.3901,
+      "eval_samples_per_second": 65.464,
+      "eval_steps_per_second": 2.158,
+      "step": 20
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5604395604395604,
+      "eval_loss": 1.1483124494552612,
+      "eval_runtime": 1.8939,
+      "eval_samples_per_second": 48.049,
+      "eval_steps_per_second": 1.584,
+      "step": 22
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 17.848636627197266,
+      "learning_rate": 9.57558796803852e-06,
+      "loss": 7.6512,
+      "step": 24
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5824175824175825,
+      "eval_loss": 1.1232386827468872,
+      "eval_runtime": 1.4211,
+      "eval_samples_per_second": 64.036,
+      "eval_steps_per_second": 2.111,
+      "step": 24
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.6043956043956044,
+      "eval_loss": 1.1041640043258667,
+      "eval_runtime": 1.9191,
+      "eval_samples_per_second": 47.417,
+      "eval_steps_per_second": 1.563,
+      "step": 26
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6153846153846154,
+      "eval_loss": 1.0891468524932861,
+      "eval_runtime": 1.6383,
+      "eval_samples_per_second": 55.544,
+      "eval_steps_per_second": 1.831,
+      "step": 28
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.6373626373626373,
+      "eval_loss": 1.0777134895324707,
+      "eval_runtime": 1.4517,
+      "eval_samples_per_second": 62.684,
+      "eval_steps_per_second": 2.067,
+      "step": 30
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.6263736263736264,
+      "eval_loss": 1.0686392784118652,
+      "eval_runtime": 1.4134,
+      "eval_samples_per_second": 64.385,
+      "eval_steps_per_second": 2.123,
+      "step": 32
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.6263736263736264,
+      "eval_loss": 1.0623126029968262,
+      "eval_runtime": 1.8859,
+      "eval_samples_per_second": 48.254,
+      "eval_steps_per_second": 1.591,
+      "step": 34
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 18.268535614013672,
+      "learning_rate": 1.2698588683881185e-06,
+      "loss": 6.5906,
+      "step": 36
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.6263736263736264,
+      "eval_loss": 1.058200478553772,
+      "eval_runtime": 1.8272,
+      "eval_samples_per_second": 49.802,
+      "eval_steps_per_second": 1.642,
+      "step": 36
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.6153846153846154,
+      "eval_loss": 1.0559276342391968,
+      "eval_runtime": 1.4242,
+      "eval_samples_per_second": 63.896,
+      "eval_steps_per_second": 2.106,
+      "step": 38
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.6153846153846154,
+      "eval_loss": 1.0549746751785278,
+      "eval_runtime": 1.4077,
+      "eval_samples_per_second": 64.643,
+      "eval_steps_per_second": 2.131,
+      "step": 40
+    },
+    {
+      "epoch": 21.0,
+      "eval_accuracy": 0.6153846153846154,
+      "eval_loss": 1.0547930002212524,
+      "eval_runtime": 2.9361,
+      "eval_samples_per_second": 30.994,
+      "eval_steps_per_second": 1.022,
+      "step": 42
+    },
+    {
+      "epoch": 21.0,
+      "step": 42,
+      "total_flos": 2.7877744402602394e+17,
+      "train_loss": 7.695562998453776,
+      "train_runtime": 380.562,
+      "train_samples_per_second": 45.028,
+      "train_steps_per_second": 0.11
+    }
+  ],
+  "logging_steps": 12,
+  "max_steps": 42,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 42,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.7877744402602394e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}