End of training

README.md

@@ -33,11 +33,11 @@ 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: 0.7837
+- Loss: 0.8253
 - Accuracy: 0.8333
-- Auc Overall: 0.875
+- Auc Overall: 0.8333
 - Auc Class 0: 1.0
-- Auc Class 1: 0.625
+- Auc Class 1: 0.5
 - Auc Class 2: 1.0
 
 ## Model description

all_results.json

@@ -1,17 +1,17 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.5,
-    "eval_auc_class_0": 0.5,
-    "eval_auc_class_1": 0.625,
-    "eval_auc_class_2": 0.625,
-    "eval_auc_overall": 0.5833333333333334,
-    "eval_loss": 1.09449303150177,
-    "eval_runtime": 0.6826,
-    "eval_samples_per_second": 8.79,
-    "eval_steps_per_second": 1.465,
-    "total_flos": 8226380544399360.0,
-    "train_loss": 0.1908613954271589,
-    "train_runtime": 147.7634,
-    "train_samples_per_second": 0.711,
-    "train_steps_per_second": 0.047
+    "epoch": 3.0,
+    "eval_accuracy": 0.8333333333333334,
+    "eval_auc_class_0": 1.0,
+    "eval_auc_class_1": 0.5,
+    "eval_auc_class_2": 1.0,
+    "eval_auc_overall": 0.8333333333333334,
+    "eval_loss": 0.8253422379493713,
+    "eval_runtime": 0.6818,
+    "eval_samples_per_second": 8.8,
+    "eval_steps_per_second": 1.467,
+    "total_flos": 3525591661885440.0,
+    "train_loss": 0.2555307348569234,
+    "train_runtime": 72.9305,
+    "train_samples_per_second": 0.617,
+    "train_steps_per_second": 0.041
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.5,
-    "eval_auc_class_0": 0.5,
-    "eval_auc_class_1": 0.625,
-    "eval_auc_class_2": 0.625,
-    "eval_auc_overall": 0.5833333333333334,
-    "eval_loss": 1.09449303150177,
-    "eval_runtime": 0.6826,
-    "eval_samples_per_second": 8.79,
-    "eval_steps_per_second": 1.465
+    "epoch": 3.0,
+    "eval_accuracy": 0.8333333333333334,
+    "eval_auc_class_0": 1.0,
+    "eval_auc_class_1": 0.5,
+    "eval_auc_class_2": 1.0,
+    "eval_auc_overall": 0.8333333333333334,
+    "eval_loss": 0.8253422379493713,
+    "eval_runtime": 0.6818,
+    "eval_samples_per_second": 8.8,
+    "eval_steps_per_second": 1.467
 }

runs/Sep04_10-43-53_data-science-wbi/events.out.tfevents.1725446753.data-science-wbi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83d469378b885fb34fb26097b31c51d29bfaacf553f32c68ba76b5cd26fd2ba1
+size 87

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 7.0,
-    "total_flos": 8226380544399360.0,
-    "train_loss": 0.1908613954271589,
-    "train_runtime": 147.7634,
-    "train_samples_per_second": 0.711,
-    "train_steps_per_second": 0.047
+    "epoch": 3.0,
+    "total_flos": 3525591661885440.0,
+    "train_loss": 0.2555307348569234,
+    "train_runtime": 72.9305,
+    "train_samples_per_second": 0.617,
+    "train_steps_per_second": 0.041
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.5,
+  "best_metric": 0.8333333333333334,
   "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-2",
-  "epoch": 7.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 7,
+  "global_step": 3,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,199 +11,95 @@
     {
       "epoch": 1.0,
       "step": 1,
-      "train_accuracy": 0.2,
-      "train_auc_class_0": 0.45999999999999996,
-      "train_auc_class_1": 0.28,
-      "train_auc_class_2": 0.38,
-      "train_auc_overall": 0.37333333333333335,
-      "train_loss": 1.2419202327728271,
-      "train_runtime": 1.6771,
-      "train_samples_per_second": 8.944,
-      "train_steps_per_second": 0.596
+      "train_accuracy": 0.6,
+      "train_auc_class_0": 0.64,
+      "train_auc_class_1": 0.46,
+      "train_auc_class_2": 0.8,
+      "train_auc_overall": 0.6333333333333334,
+      "train_loss": 1.0219470262527466,
+      "train_runtime": 1.8345,
+      "train_samples_per_second": 8.177,
+      "train_steps_per_second": 0.545
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.0,
-      "eval_auc_class_0": 0.5,
-      "eval_auc_class_1": 0.25,
-      "eval_auc_class_2": 0.5,
-      "eval_auc_overall": 0.4166666666666667,
-      "eval_loss": 1.1770787239074707,
-      "eval_runtime": 0.97,
-      "eval_samples_per_second": 6.185,
-      "eval_steps_per_second": 1.031,
+      "eval_accuracy": 0.6666666666666666,
+      "eval_auc_class_0": 1.0,
+      "eval_auc_class_1": 0.5,
+      "eval_auc_class_2": 0.875,
+      "eval_auc_overall": 0.7916666666666666,
+      "eval_loss": 0.9240671992301941,
+      "eval_runtime": 0.7631,
+      "eval_samples_per_second": 7.863,
+      "eval_steps_per_second": 1.31,
       "step": 1
     },
     {
       "epoch": 2.0,
       "step": 2,
-      "train_accuracy": 0.6666666666666666,
-      "train_auc_class_0": 0.82,
-      "train_auc_class_1": 0.92,
-      "train_auc_class_2": 0.86,
-      "train_auc_overall": 0.8666666666666667,
-      "train_loss": 0.8910435438156128,
-      "train_runtime": 1.6903,
-      "train_samples_per_second": 8.874,
-      "train_steps_per_second": 0.592
+      "train_accuracy": 0.8,
+      "train_auc_class_0": 1.0,
+      "train_auc_class_1": 0.96,
+      "train_auc_class_2": 1.0,
+      "train_auc_overall": 0.9866666666666667,
+      "train_loss": 0.6553783416748047,
+      "train_runtime": 1.6643,
+      "train_samples_per_second": 9.013,
+      "train_steps_per_second": 0.601
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.5,
-      "eval_auc_class_1": 0.625,
-      "eval_auc_class_2": 0.625,
-      "eval_auc_overall": 0.5833333333333334,
-      "eval_loss": 1.09449303150177,
-      "eval_runtime": 0.7379,
-      "eval_samples_per_second": 8.132,
-      "eval_steps_per_second": 1.355,
+      "eval_accuracy": 0.8333333333333334,
+      "eval_auc_class_0": 1.0,
+      "eval_auc_class_1": 0.5,
+      "eval_auc_class_2": 1.0,
+      "eval_auc_overall": 0.8333333333333334,
+      "eval_loss": 0.8253422379493713,
+      "eval_runtime": 0.6909,
+      "eval_samples_per_second": 8.685,
+      "eval_steps_per_second": 1.447,
       "step": 2
     },
     {
       "epoch": 3.0,
       "step": 3,
       "train_accuracy": 0.8666666666666667,
-      "train_auc_class_0": 0.98,
+      "train_auc_class_0": 0.96,
       "train_auc_class_1": 1.0,
       "train_auc_class_2": 0.98,
-      "train_auc_overall": 0.9866666666666667,
-      "train_loss": 0.6887053847312927,
-      "train_runtime": 1.7287,
-      "train_samples_per_second": 8.677,
-      "train_steps_per_second": 0.578
+      "train_auc_overall": 0.98,
+      "train_loss": 0.7010973691940308,
+      "train_runtime": 1.7627,
+      "train_samples_per_second": 8.51,
+      "train_steps_per_second": 0.567
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.5,
-      "eval_auc_class_1": 0.75,
-      "eval_auc_class_2": 0.75,
-      "eval_auc_overall": 0.6666666666666666,
-      "eval_loss": 1.030716896057129,
-      "eval_runtime": 0.7962,
-      "eval_samples_per_second": 7.536,
-      "eval_steps_per_second": 1.256,
-      "step": 3
-    },
-    {
-      "epoch": 4.0,
-      "step": 4,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.5273771286010742,
-      "train_runtime": 1.7311,
-      "train_samples_per_second": 8.665,
-      "train_steps_per_second": 0.578
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.75,
-      "eval_auc_class_1": 0.875,
-      "eval_auc_class_2": 0.875,
-      "eval_auc_overall": 0.8333333333333334,
-      "eval_loss": 0.9745334982872009,
-      "eval_runtime": 0.708,
-      "eval_samples_per_second": 8.474,
-      "eval_steps_per_second": 1.412,
-      "step": 4
-    },
-    {
-      "epoch": 5.0,
-      "step": 5,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.4640212059020996,
-      "train_runtime": 1.7287,
-      "train_samples_per_second": 8.677,
-      "train_steps_per_second": 0.578
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.75,
-      "eval_auc_class_1": 0.875,
-      "eval_auc_class_2": 0.875,
-      "eval_auc_overall": 0.8333333333333334,
-      "eval_loss": 0.928966224193573,
-      "eval_runtime": 0.7026,
-      "eval_samples_per_second": 8.54,
-      "eval_steps_per_second": 1.423,
-      "step": 5
-    },
-    {
-      "epoch": 6.0,
-      "step": 6,
-      "train_accuracy": 0.9333333333333333,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.4194321036338806,
-      "train_runtime": 1.7268,
-      "train_samples_per_second": 8.687,
-      "train_steps_per_second": 0.579
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.75,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 0.875,
+      "eval_accuracy": 0.8333333333333334,
+      "eval_auc_class_0": 1.0,
+      "eval_auc_class_1": 0.625,
+      "eval_auc_class_2": 1.0,
       "eval_auc_overall": 0.875,
-      "eval_loss": 0.8979067206382751,
-      "eval_runtime": 0.6949,
-      "eval_samples_per_second": 8.634,
-      "eval_steps_per_second": 1.439,
-      "step": 6
-    },
-    {
-      "epoch": 7.0,
-      "step": 7,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.402330219745636,
-      "train_runtime": 1.8612,
-      "train_samples_per_second": 8.059,
-      "train_steps_per_second": 0.537
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.875,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 0.875,
-      "eval_auc_overall": 0.9166666666666666,
-      "eval_loss": 0.8809628486633301,
-      "eval_runtime": 0.7654,
-      "eval_samples_per_second": 7.839,
-      "eval_steps_per_second": 1.306,
-      "step": 7
+      "eval_loss": 0.7836728096008301,
+      "eval_runtime": 0.7323,
+      "eval_samples_per_second": 8.193,
+      "eval_steps_per_second": 1.366,
+      "step": 3
     },
     {
-      "epoch": 7.0,
-      "step": 7,
-      "total_flos": 8226380544399360.0,
-      "train_loss": 0.1908613954271589,
-      "train_runtime": 147.7634,
-      "train_samples_per_second": 0.711,
-      "train_steps_per_second": 0.047
+      "epoch": 3.0,
+      "step": 3,
+      "total_flos": 3525591661885440.0,
+      "train_loss": 0.2555307348569234,
+      "train_runtime": 72.9305,
+      "train_samples_per_second": 0.617,
+      "train_steps_per_second": 0.041
     }
   ],
   "logging_steps": 10,
-  "max_steps": 7,
+  "max_steps": 3,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -217,7 +113,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8226380544399360.0,
+  "total_flos": 3525591661885440.0,
   "train_batch_size": 128,
   "trial_name": null,
   "trial_params": null