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: 0.9945
+- Loss: 1.0560
 - Accuracy: 0.5
-- Auc Overall: 0.625
+- Auc Overall: 0.5833
 - Auc Class 0: 0.25
-- Auc Class 1: 0.625
+- Auc Class 1: 0.5
 - Auc Class 2: 1.0
 
 ## Model description

all_results.json

@@ -1,22 +1,22 @@
 {
     "epoch": 3.0,
     "eval_accuracy": 0.5,
-    "eval_auc_class_0": 0.625,
-    "eval_auc_class_1": 0.375,
-    "eval_auc_class_2": 0.0,
-    "eval_auc_overall": 0.3333333333333333,
-    "eval_loss": 1.2536860704421997,
-    "eval_runtime": 0.7286,
-    "eval_samples_per_second": 8.234,
-    "eval_steps_per_second": 1.372,
+    "eval_auc_class_0": 0.25,
+    "eval_auc_class_1": 0.5,
+    "eval_auc_class_2": 1.0,
+    "eval_auc_overall": 0.5833333333333334,
+    "eval_loss": 1.0559865236282349,
+    "eval_runtime": 0.6886,
+    "eval_samples_per_second": 8.713,
+    "eval_steps_per_second": 1.452,
     "total_flos": 3525591661885440.0,
     "train_accuracy": 0.6,
     "train_auc_class_0": 0.7,
     "train_auc_class_1": 0.96,
     "train_auc_class_2": 0.8200000000000001,
     "train_auc_overall": 0.8266666666666667,
-    "train_loss": 0.9045717120170593,
-    "train_runtime": 1.6579,
-    "train_samples_per_second": 9.047,
-    "train_steps_per_second": 0.603
+    "train_loss": 0.2689722975095113,
+    "train_runtime": 72.4115,
+    "train_samples_per_second": 0.621,
+    "train_steps_per_second": 0.041
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 3.0,
     "eval_accuracy": 0.5,
-    "eval_auc_class_0": 0.625,
-    "eval_auc_class_1": 0.375,
-    "eval_auc_class_2": 0.0,
-    "eval_auc_overall": 0.3333333333333333,
-    "eval_loss": 1.2536860704421997,
-    "eval_runtime": 0.7286,
-    "eval_samples_per_second": 8.234,
-    "eval_steps_per_second": 1.372
+    "eval_auc_class_0": 0.25,
+    "eval_auc_class_1": 0.5,
+    "eval_auc_class_2": 1.0,
+    "eval_auc_overall": 0.5833333333333334,
+    "eval_loss": 1.0559865236282349,
+    "eval_runtime": 0.6886,
+    "eval_samples_per_second": 8.713,
+    "eval_steps_per_second": 1.452
 }

runs/Sep04_11-10-34_data-science-wbi/events.out.tfevents.1725448310.data-science-wbi

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec8bbc02d38802115efbb1b5d01fc06563d9f6024ad375059355240d4c155b60
-size 94
+oid sha256:a827e650a02355c170a7ad5ee99ff98fb860c1a4ca21e31e77a2b98e81d151c7
+size 636

train_results.json

@@ -1,12 +1,8 @@
 {
     "epoch": 3.0,
-    "train_accuracy": 0.6,
-    "train_auc_class_0": 0.7,
-    "train_auc_class_1": 0.96,
-    "train_auc_class_2": 0.8200000000000001,
-    "train_auc_overall": 0.8266666666666667,
-    "train_loss": 0.9045717120170593,
-    "train_runtime": 1.6579,
-    "train_samples_per_second": 9.047,
-    "train_steps_per_second": 0.603
+    "total_flos": 3525591661885440.0,
+    "train_loss": 0.2689722975095113,
+    "train_runtime": 72.4115,
+    "train_samples_per_second": 0.621,
+    "train_steps_per_second": 0.041
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.5,
-  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-1",
+  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-2",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 3,
@@ -11,102 +11,102 @@
     {
       "epoch": 1.0,
       "step": 1,
-      "train_accuracy": 0.3333333333333333,
-      "train_auc_class_0": 0.52,
-      "train_auc_class_1": 0.5800000000000001,
-      "train_auc_class_2": 0.22,
-      "train_auc_overall": 0.44,
-      "train_loss": 1.3131554126739502,
-      "train_runtime": 1.6412,
-      "train_samples_per_second": 9.14,
-      "train_steps_per_second": 0.609
+      "train_accuracy": 0.5333333333333333,
+      "train_auc_class_0": 0.34,
+      "train_auc_class_1": 0.6599999999999999,
+      "train_auc_class_2": 0.5599999999999999,
+      "train_auc_overall": 0.52,
+      "train_loss": 1.1574488878250122,
+      "train_runtime": 1.6568,
+      "train_samples_per_second": 9.054,
+      "train_steps_per_second": 0.604
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.625,
+      "eval_accuracy": 0.3333333333333333,
+      "eval_auc_class_0": 0.25,
       "eval_auc_class_1": 0.375,
-      "eval_auc_class_2": 0.0,
-      "eval_auc_overall": 0.3333333333333333,
-      "eval_loss": 1.2536860704421997,
-      "eval_runtime": 0.6979,
-      "eval_samples_per_second": 8.597,
-      "eval_steps_per_second": 1.433,
+      "eval_auc_class_2": 1.0,
+      "eval_auc_overall": 0.5416666666666666,
+      "eval_loss": 1.2032381296157837,
+      "eval_runtime": 0.6994,
+      "eval_samples_per_second": 8.579,
+      "eval_steps_per_second": 1.43,
       "step": 1
     },
     {
       "epoch": 2.0,
       "step": 2,
-      "train_accuracy": 0.6666666666666666,
-      "train_auc_class_0": 0.7799999999999999,
-      "train_auc_class_1": 0.8799999999999999,
-      "train_auc_class_2": 0.64,
-      "train_auc_overall": 0.7666666666666666,
-      "train_loss": 0.9476243853569031,
-      "train_runtime": 1.6529,
-      "train_samples_per_second": 9.075,
-      "train_steps_per_second": 0.605
+      "train_accuracy": 0.6,
+      "train_auc_class_0": 0.5,
+      "train_auc_class_1": 0.92,
+      "train_auc_class_2": 0.8,
+      "train_auc_overall": 0.7399999999999999,
+      "train_loss": 0.9586163759231567,
+      "train_runtime": 1.6488,
+      "train_samples_per_second": 9.097,
+      "train_steps_per_second": 0.606
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.3333333333333333,
-      "eval_auc_class_0": 0.625,
-      "eval_auc_class_1": 0.75,
-      "eval_auc_class_2": 0.125,
-      "eval_auc_overall": 0.5,
-      "eval_loss": 1.089605689048767,
-      "eval_runtime": 0.7552,
-      "eval_samples_per_second": 7.945,
-      "eval_steps_per_second": 1.324,
+      "eval_accuracy": 0.5,
+      "eval_auc_class_0": 0.25,
+      "eval_auc_class_1": 0.5,
+      "eval_auc_class_2": 1.0,
+      "eval_auc_overall": 0.5833333333333334,
+      "eval_loss": 1.0559865236282349,
+      "eval_runtime": 0.6879,
+      "eval_samples_per_second": 8.723,
+      "eval_steps_per_second": 1.454,
       "step": 2
     },
     {
       "epoch": 3.0,
       "step": 3,
-      "train_accuracy": 0.8,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 0.98,
-      "train_auc_class_2": 0.9199999999999999,
-      "train_auc_overall": 0.9666666666666667,
-      "train_loss": 0.7161972522735596,
-      "train_runtime": 1.7658,
-      "train_samples_per_second": 8.495,
-      "train_steps_per_second": 0.566
+      "train_accuracy": 0.6666666666666666,
+      "train_auc_class_0": 0.7399999999999999,
+      "train_auc_class_1": 0.8800000000000001,
+      "train_auc_class_2": 0.8600000000000001,
+      "train_auc_overall": 0.8266666666666668,
+      "train_loss": 0.8743086457252502,
+      "train_runtime": 1.6839,
+      "train_samples_per_second": 8.908,
+      "train_steps_per_second": 0.594
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.3333333333333333,
-      "eval_auc_class_0": 0.625,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 0.25,
+      "eval_accuracy": 0.5,
+      "eval_auc_class_0": 0.25,
+      "eval_auc_class_1": 0.625,
+      "eval_auc_class_2": 1.0,
       "eval_auc_overall": 0.625,
-      "eval_loss": 1.0246309041976929,
-      "eval_runtime": 0.7191,
-      "eval_samples_per_second": 8.344,
-      "eval_steps_per_second": 1.391,
+      "eval_loss": 0.9944595694541931,
+      "eval_runtime": 0.6872,
+      "eval_samples_per_second": 8.73,
+      "eval_steps_per_second": 1.455,
       "step": 3
     },
     {
       "epoch": 3.0,
       "step": 3,
       "total_flos": 3525591661885440.0,
-      "train_loss": 0.2987220684687297,
-      "train_runtime": 73.0126,
-      "train_samples_per_second": 0.616,
+      "train_loss": 0.2689722975095113,
+      "train_runtime": 72.4115,
+      "train_samples_per_second": 0.621,
       "train_steps_per_second": 0.041
     },
     {
       "epoch": 3.0,
       "step": 3,
-      "train_accuracy": 0.13333333333333333,
-      "train_auc_class_0": 0.39999999999999997,
-      "train_auc_class_1": 0.6000000000000001,
-      "train_auc_class_2": 0.2,
-      "train_auc_overall": 0.39999999999999997,
-      "train_loss": 1.3039982318878174,
-      "train_runtime": 1.7052,
-      "train_samples_per_second": 8.797,
-      "train_steps_per_second": 0.586
+      "train_accuracy": 0.6,
+      "train_auc_class_0": 0.7,
+      "train_auc_class_1": 0.96,
+      "train_auc_class_2": 0.8200000000000001,
+      "train_auc_overall": 0.8266666666666667,
+      "train_loss": 0.9045717120170593,
+      "train_runtime": 1.6579,
+      "train_samples_per_second": 9.047,
+      "train_steps_per_second": 0.603
     }
   ],
   "logging_steps": 10,