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/swinv2-base-patch4-window8-256](https://huggingface.co/microsoft/swinv2-base-patch4-window8-256) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4871
+- Loss: 0.5212
 - Accuracy: 0.8140
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 9.6,
+    "eval_accuracy": 0.813953488372093,
+    "eval_loss": 0.5212067365646362,
+    "eval_runtime": 0.5228,
+    "eval_samples_per_second": 82.249,
+    "eval_steps_per_second": 21.041,
+    "total_flos": 9.748371434294477e+16,
+    "train_loss": 0.5928216457366944,
+    "train_runtime": 55.1953,
+    "train_samples_per_second": 17.936,
+    "train_steps_per_second": 1.087
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.6,
+    "eval_accuracy": 0.813953488372093,
+    "eval_loss": 0.5212067365646362,
+    "eval_runtime": 0.5228,
+    "eval_samples_per_second": 82.249,
+    "eval_steps_per_second": 21.041
+}

runs/Aug10_15-28-49_EMIMDGXA100GPU1/events.out.tfevents.1723318210.EMIMDGXA100GPU1.4135509.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04dfd99e87cbcca127905b96619af64653a4c0266182d1e2c924a19bad7ab4f0
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.6,
+    "total_flos": 9.748371434294477e+16,
+    "train_loss": 0.5928216457366944,
+    "train_runtime": 55.1953,
+    "train_samples_per_second": 17.936,
+    "train_steps_per_second": 1.087
+}

trainer_state.json

@@ -0,0 +1,174 @@
+{
+  "best_metric": 0.813953488372093,
+  "best_model_checkpoint": "swinv2-base-patch4-window8-256/checkpoint-12",
+  "epoch": 9.6,
+  "eval_steps": 500,
+  "global_step": 60,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.96,
+      "eval_accuracy": 0.7906976744186046,
+      "eval_loss": 0.602724552154541,
+      "eval_runtime": 0.4667,
+      "eval_samples_per_second": 92.138,
+      "eval_steps_per_second": 23.57,
+      "step": 6
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 5.375301837921143,
+      "learning_rate": 4.62962962962963e-06,
+      "loss": 0.6492,
+      "step": 10
+    },
+    {
+      "epoch": 1.92,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.5212067365646362,
+      "eval_runtime": 0.478,
+      "eval_samples_per_second": 89.95,
+      "eval_steps_per_second": 23.011,
+      "step": 12
+    },
+    {
+      "epoch": 2.88,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.4939202666282654,
+      "eval_runtime": 0.4952,
+      "eval_samples_per_second": 86.835,
+      "eval_steps_per_second": 22.214,
+      "step": 18
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 7.962141036987305,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.5773,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.48294201493263245,
+      "eval_runtime": 0.4872,
+      "eval_samples_per_second": 88.252,
+      "eval_steps_per_second": 22.576,
+      "step": 25
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 7.068347930908203,
+      "learning_rate": 2.7777777777777783e-06,
+      "loss": 0.6313,
+      "step": 30
+    },
+    {
+      "epoch": 4.96,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.48332729935646057,
+      "eval_runtime": 0.5067,
+      "eval_samples_per_second": 84.87,
+      "eval_steps_per_second": 21.711,
+      "step": 31
+    },
+    {
+      "epoch": 5.92,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.4872807264328003,
+      "eval_runtime": 0.5598,
+      "eval_samples_per_second": 76.807,
+      "eval_steps_per_second": 19.648,
+      "step": 37
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 7.162075042724609,
+      "learning_rate": 1.8518518518518519e-06,
+      "loss": 0.5665,
+      "step": 40
+    },
+    {
+      "epoch": 6.88,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.48763200640678406,
+      "eval_runtime": 0.4695,
+      "eval_samples_per_second": 91.588,
+      "eval_steps_per_second": 23.43,
+      "step": 43
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.575540542602539,
+      "learning_rate": 9.259259259259259e-07,
+      "loss": 0.5615,
+      "step": 50
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.4861143231391907,
+      "eval_runtime": 0.4886,
+      "eval_samples_per_second": 88.012,
+      "eval_steps_per_second": 22.515,
+      "step": 50
+    },
+    {
+      "epoch": 8.96,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.487347811460495,
+      "eval_runtime": 0.4891,
+      "eval_samples_per_second": 87.918,
+      "eval_steps_per_second": 22.491,
+      "step": 56
+    },
+    {
+      "epoch": 9.6,
+      "grad_norm": 5.972108840942383,
+      "learning_rate": 0.0,
+      "loss": 0.5712,
+      "step": 60
+    },
+    {
+      "epoch": 9.6,
+      "eval_accuracy": 0.813953488372093,
+      "eval_loss": 0.48711302876472473,
+      "eval_runtime": 0.5016,
+      "eval_samples_per_second": 85.725,
+      "eval_steps_per_second": 21.93,
+      "step": 60
+    },
+    {
+      "epoch": 9.6,
+      "step": 60,
+      "total_flos": 9.748371434294477e+16,
+      "train_loss": 0.5928216457366944,
+      "train_runtime": 55.1953,
+      "train_samples_per_second": 17.936,
+      "train_steps_per_second": 1.087
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 60,
+  "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": 9.748371434294477e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}