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 [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1881
+- Loss: 0.1890
 - Accuracy: 0.9366
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 9.68421052631579,
+    "eval_accuracy": 0.9365918097754293,
+    "eval_loss": 0.18896903097629547,
+    "eval_runtime": 4.222,
+    "eval_samples_per_second": 179.298,
+    "eval_steps_per_second": 5.684,
+    "total_flos": 7.494072602457047e+17,
+    "train_loss": 0.7879104811212291,
+    "train_runtime": 284.9792,
+    "train_samples_per_second": 106.253,
+    "train_steps_per_second": 0.807
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.68421052631579,
+    "eval_accuracy": 0.9365918097754293,
+    "eval_loss": 0.18896903097629547,
+    "eval_runtime": 4.222,
+    "eval_samples_per_second": 179.298,
+    "eval_steps_per_second": 5.684
+}

runs/Jun07_11-44-27_deb7c8f53bdf/events.out.tfevents.1717760964.deb7c8f53bdf.1038.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fddf0992b85a13ffc1b3f8f2d0721ddd32274349000f124c04ec0f28d4295f1
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.68421052631579,
+    "total_flos": 7.494072602457047e+17,
+    "train_loss": 0.7879104811212291,
+    "train_runtime": 284.9792,
+    "train_samples_per_second": 106.253,
+    "train_steps_per_second": 0.807
+}

trainer_state.json

@@ -0,0 +1,293 @@
+{
+  "best_metric": 0.9365918097754293,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-landscape/checkpoint-213",
+  "epoch": 9.68421052631579,
+  "eval_steps": 500,
+  "global_step": 230,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 70.45781707763672,
+      "learning_rate": 2.173913043478261e-05,
+      "loss": 7.8634,
+      "step": 10
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 27.836477279663086,
+      "learning_rate": 4.347826086956522e-05,
+      "loss": 3.6193,
+      "step": 20
+    },
+    {
+      "epoch": 0.968421052631579,
+      "eval_accuracy": 0.6618229854689565,
+      "eval_loss": 1.0203161239624023,
+      "eval_runtime": 3.846,
+      "eval_samples_per_second": 196.827,
+      "eval_steps_per_second": 6.24,
+      "step": 23
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 14.402323722839355,
+      "learning_rate": 4.830917874396135e-05,
+      "loss": 1.2754,
+      "step": 30
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 17.051239013671875,
+      "learning_rate": 4.589371980676328e-05,
+      "loss": 0.5876,
+      "step": 40
+    },
+    {
+      "epoch": 1.9789473684210526,
+      "eval_accuracy": 0.8903566710700133,
+      "eval_loss": 0.3358216881752014,
+      "eval_runtime": 3.7597,
+      "eval_samples_per_second": 201.344,
+      "eval_steps_per_second": 6.383,
+      "step": 47
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 12.79457950592041,
+      "learning_rate": 4.347826086956522e-05,
+      "loss": 0.3806,
+      "step": 50
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 7.795621395111084,
+      "learning_rate": 4.106280193236715e-05,
+      "loss": 0.3839,
+      "step": 60
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 8.323637008666992,
+      "learning_rate": 3.864734299516908e-05,
+      "loss": 0.3585,
+      "step": 70
+    },
+    {
+      "epoch": 2.9894736842105263,
+      "eval_accuracy": 0.9207397622192867,
+      "eval_loss": 0.2493603378534317,
+      "eval_runtime": 3.5421,
+      "eval_samples_per_second": 213.718,
+      "eval_steps_per_second": 6.776,
+      "step": 71
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 5.737844944000244,
+      "learning_rate": 3.6231884057971014e-05,
+      "loss": 0.2795,
+      "step": 80
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 5.775590896606445,
+      "learning_rate": 3.381642512077295e-05,
+      "loss": 0.2906,
+      "step": 90
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9207397622192867,
+      "eval_loss": 0.23325380682945251,
+      "eval_runtime": 4.0394,
+      "eval_samples_per_second": 187.403,
+      "eval_steps_per_second": 5.941,
+      "step": 95
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 7.07818603515625,
+      "learning_rate": 3.140096618357488e-05,
+      "loss": 0.3048,
+      "step": 100
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 7.080397605895996,
+      "learning_rate": 2.8985507246376814e-05,
+      "loss": 0.2591,
+      "step": 110
+    },
+    {
+      "epoch": 4.968421052631579,
+      "eval_accuracy": 0.9313077939233818,
+      "eval_loss": 0.2077917903661728,
+      "eval_runtime": 3.7706,
+      "eval_samples_per_second": 200.761,
+      "eval_steps_per_second": 6.365,
+      "step": 118
+    },
+    {
+      "epoch": 5.052631578947368,
+      "grad_norm": 5.3396124839782715,
+      "learning_rate": 2.6570048309178748e-05,
+      "loss": 0.2752,
+      "step": 120
+    },
+    {
+      "epoch": 5.473684210526316,
+      "grad_norm": 5.555719375610352,
+      "learning_rate": 2.4154589371980676e-05,
+      "loss": 0.2328,
+      "step": 130
+    },
+    {
+      "epoch": 5.894736842105263,
+      "grad_norm": 5.745518684387207,
+      "learning_rate": 2.173913043478261e-05,
+      "loss": 0.2458,
+      "step": 140
+    },
+    {
+      "epoch": 5.978947368421053,
+      "eval_accuracy": 0.9233817701453104,
+      "eval_loss": 0.22294138371944427,
+      "eval_runtime": 3.6849,
+      "eval_samples_per_second": 205.433,
+      "eval_steps_per_second": 6.513,
+      "step": 142
+    },
+    {
+      "epoch": 6.315789473684211,
+      "grad_norm": 6.548435688018799,
+      "learning_rate": 1.932367149758454e-05,
+      "loss": 0.2252,
+      "step": 150
+    },
+    {
+      "epoch": 6.7368421052631575,
+      "grad_norm": 6.568725109100342,
+      "learning_rate": 1.6908212560386476e-05,
+      "loss": 0.1936,
+      "step": 160
+    },
+    {
+      "epoch": 6.989473684210527,
+      "eval_accuracy": 0.9220607661822986,
+      "eval_loss": 0.20947237312793732,
+      "eval_runtime": 3.5585,
+      "eval_samples_per_second": 212.729,
+      "eval_steps_per_second": 6.744,
+      "step": 166
+    },
+    {
+      "epoch": 7.157894736842105,
+      "grad_norm": 5.143336772918701,
+      "learning_rate": 1.4492753623188407e-05,
+      "loss": 0.2081,
+      "step": 170
+    },
+    {
+      "epoch": 7.578947368421053,
+      "grad_norm": 7.358247756958008,
+      "learning_rate": 1.2077294685990338e-05,
+      "loss": 0.2262,
+      "step": 180
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 5.925826549530029,
+      "learning_rate": 9.66183574879227e-06,
+      "loss": 0.1765,
+      "step": 190
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9247027741083224,
+      "eval_loss": 0.19408579170703888,
+      "eval_runtime": 3.8932,
+      "eval_samples_per_second": 194.44,
+      "eval_steps_per_second": 6.165,
+      "step": 190
+    },
+    {
+      "epoch": 8.421052631578947,
+      "grad_norm": 4.0063252449035645,
+      "learning_rate": 7.246376811594203e-06,
+      "loss": 0.1819,
+      "step": 200
+    },
+    {
+      "epoch": 8.842105263157894,
+      "grad_norm": 4.924877166748047,
+      "learning_rate": 4.830917874396135e-06,
+      "loss": 0.1915,
+      "step": 210
+    },
+    {
+      "epoch": 8.968421052631578,
+      "eval_accuracy": 0.9365918097754293,
+      "eval_loss": 0.18896903097629547,
+      "eval_runtime": 4.0115,
+      "eval_samples_per_second": 188.708,
+      "eval_steps_per_second": 5.983,
+      "step": 213
+    },
+    {
+      "epoch": 9.263157894736842,
+      "grad_norm": 5.235773086547852,
+      "learning_rate": 2.4154589371980677e-06,
+      "loss": 0.1798,
+      "step": 220
+    },
+    {
+      "epoch": 9.68421052631579,
+      "grad_norm": 4.68380880355835,
+      "learning_rate": 0.0,
+      "loss": 0.1824,
+      "step": 230
+    },
+    {
+      "epoch": 9.68421052631579,
+      "eval_accuracy": 0.9365918097754293,
+      "eval_loss": 0.18814285099506378,
+      "eval_runtime": 3.7835,
+      "eval_samples_per_second": 200.08,
+      "eval_steps_per_second": 6.343,
+      "step": 230
+    },
+    {
+      "epoch": 9.68421052631579,
+      "step": 230,
+      "total_flos": 7.494072602457047e+17,
+      "train_loss": 0.7879104811212291,
+      "train_runtime": 284.9792,
+      "train_samples_per_second": 106.253,
+      "train_steps_per_second": 0.807
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 230,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "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": 7.494072602457047e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}