End of training

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 2.984771573604061,
+    "eval_accuracy": 0.9772727272727273,
+    "eval_loss": 0.05173008143901825,
+    "eval_runtime": 1.5518,
+    "eval_samples_per_second": 113.415,
+    "eval_steps_per_second": 14.177,
+    "total_flos": 1.1692257218946662e+17,
+    "train_loss": 0.2702838227051456,
+    "train_runtime": 99.2204,
+    "train_samples_per_second": 47.651,
+    "train_steps_per_second": 1.482
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.984771573604061,
+    "eval_accuracy": 0.9772727272727273,
+    "eval_loss": 0.05173008143901825,
+    "eval_runtime": 1.5518,
+    "eval_samples_per_second": 113.415,
+    "eval_steps_per_second": 14.177
+}

runs/May04_04-06-10_f7072774099d/events.out.tfevents.1714795720.f7072774099d.1409.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f1ad02b94ba2eebc32c596556b896fc099366684925649ff2c632d473454eec
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.984771573604061,
+    "total_flos": 1.1692257218946662e+17,
+    "train_loss": 0.2702838227051456,
+    "train_runtime": 99.2204,
+    "train_samples_per_second": 47.651,
+    "train_steps_per_second": 1.482
+}

trainer_state.json

@@ -0,0 +1,155 @@
+{
+  "best_metric": 0.9772727272727273,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-batch8/checkpoint-147",
+  "epoch": 2.984771573604061,
+  "eval_steps": 500,
+  "global_step": 147,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.20304568527918782,
+      "grad_norm": 9.754337310791016,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.6886,
+      "step": 10
+    },
+    {
+      "epoch": 0.40609137055837563,
+      "grad_norm": 13.60398006439209,
+      "learning_rate": 4.810606060606061e-05,
+      "loss": 0.5524,
+      "step": 20
+    },
+    {
+      "epoch": 0.6091370558375635,
+      "grad_norm": 13.017239570617676,
+      "learning_rate": 4.431818181818182e-05,
+      "loss": 0.3516,
+      "step": 30
+    },
+    {
+      "epoch": 0.8121827411167513,
+      "grad_norm": 10.370221138000488,
+      "learning_rate": 4.053030303030303e-05,
+      "loss": 0.2333,
+      "step": 40
+    },
+    {
+      "epoch": 0.9949238578680203,
+      "eval_accuracy": 0.9659090909090909,
+      "eval_loss": 0.07651708275079727,
+      "eval_runtime": 1.4025,
+      "eval_samples_per_second": 125.494,
+      "eval_steps_per_second": 15.687,
+      "step": 49
+    },
+    {
+      "epoch": 1.015228426395939,
+      "grad_norm": 22.6178035736084,
+      "learning_rate": 3.6742424242424246e-05,
+      "loss": 0.3668,
+      "step": 50
+    },
+    {
+      "epoch": 1.218274111675127,
+      "grad_norm": 20.655942916870117,
+      "learning_rate": 3.295454545454545e-05,
+      "loss": 0.2372,
+      "step": 60
+    },
+    {
+      "epoch": 1.4213197969543148,
+      "grad_norm": 21.389158248901367,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.1728,
+      "step": 70
+    },
+    {
+      "epoch": 1.6243654822335025,
+      "grad_norm": 7.201972961425781,
+      "learning_rate": 2.537878787878788e-05,
+      "loss": 0.1676,
+      "step": 80
+    },
+    {
+      "epoch": 1.8274111675126905,
+      "grad_norm": 15.233765602111816,
+      "learning_rate": 2.1590909090909093e-05,
+      "loss": 0.2171,
+      "step": 90
+    },
+    {
+      "epoch": 1.9898477157360406,
+      "eval_accuracy": 0.9715909090909091,
+      "eval_loss": 0.07618988305330276,
+      "eval_runtime": 1.8486,
+      "eval_samples_per_second": 95.208,
+      "eval_steps_per_second": 11.901,
+      "step": 98
+    },
+    {
+      "epoch": 2.030456852791878,
+      "grad_norm": 22.444684982299805,
+      "learning_rate": 1.7803030303030303e-05,
+      "loss": 0.2291,
+      "step": 100
+    },
+    {
+      "epoch": 2.233502538071066,
+      "grad_norm": 17.800857543945312,
+      "learning_rate": 1.4015151515151515e-05,
+      "loss": 0.2027,
+      "step": 110
+    },
+    {
+      "epoch": 2.436548223350254,
+      "grad_norm": 8.930273056030273,
+      "learning_rate": 1.0227272727272729e-05,
+      "loss": 0.1624,
+      "step": 120
+    },
+    {
+      "epoch": 2.6395939086294415,
+      "grad_norm": 2.2625389099121094,
+      "learning_rate": 6.43939393939394e-06,
+      "loss": 0.1362,
+      "step": 130
+    },
+    {
+      "epoch": 2.8426395939086295,
+      "grad_norm": 17.570310592651367,
+      "learning_rate": 2.651515151515152e-06,
+      "loss": 0.1713,
+      "step": 140
+    },
+    {
+      "epoch": 2.984771573604061,
+      "eval_accuracy": 0.9772727272727273,
+      "eval_loss": 0.05173008143901825,
+      "eval_runtime": 1.4237,
+      "eval_samples_per_second": 123.622,
+      "eval_steps_per_second": 15.453,
+      "step": 147
+    },
+    {
+      "epoch": 2.984771573604061,
+      "step": 147,
+      "total_flos": 1.1692257218946662e+17,
+      "train_loss": 0.2702838227051456,
+      "train_runtime": 99.2204,
+      "train_samples_per_second": 47.651,
+      "train_steps_per_second": 1.482
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 147,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.1692257218946662e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}