End of training

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8705334023821854
+      value: 0.8720870015535992
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,9 +33,9 @@ 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: 0.3173
-- Accuracy: 0.8705
-- Auc: 0.9550
+- Loss: 0.3245
+- Accuracy: 0.8721
+- Auc: 0.9534
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 6.885245901639344,
-    "eval_accuracy": 0.8824443293630243,
-    "eval_auc": 0.957557852599524,
-    "eval_loss": 0.30561181902885437,
-    "eval_runtime": 203.3286,
-    "eval_samples_per_second": 9.497,
+    "eval_accuracy": 0.8720870015535992,
+    "eval_auc": 0.9534343762433686,
+    "eval_loss": 0.3244781792163849,
+    "eval_runtime": 202.4915,
+    "eval_samples_per_second": 9.536,
     "eval_steps_per_second": 0.079,
     "total_flos": 4.1785312376666235e+18,
-    "train_loss": 0.38780884515671504,
-    "train_runtime": 6680.4521,
-    "train_samples_per_second": 8.107,
-    "train_steps_per_second": 0.016
+    "train_loss": 0.3947783969697498,
+    "train_runtime": 11418.3292,
+    "train_samples_per_second": 4.743,
+    "train_steps_per_second": 0.009
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 6.885245901639344,
-    "eval_accuracy": 0.8824443293630243,
-    "eval_auc": 0.957557852599524,
-    "eval_loss": 0.30561181902885437,
-    "eval_runtime": 203.3286,
-    "eval_samples_per_second": 9.497,
+    "eval_accuracy": 0.8720870015535992,
+    "eval_auc": 0.9534343762433686,
+    "eval_loss": 0.3244781792163849,
+    "eval_runtime": 202.4915,
+    "eval_samples_per_second": 9.536,
     "eval_steps_per_second": 0.079
 }

runs/Aug30_06-47-56_data-science-wbi/events.out.tfevents.1725012113.data-science-wbi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddb579b7d0e441055662d315ee540d1acd11d18b5a632a7e14b0abd344fc41f2
+size 87

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.885245901639344,
     "total_flos": 4.1785312376666235e+18,
-    "train_loss": 0.38780884515671504,
-    "train_runtime": 6680.4521,
-    "train_samples_per_second": 8.107,
-    "train_steps_per_second": 0.016
+    "train_loss": 0.3947783969697498,
+    "train_runtime": 11418.3292,
+    "train_samples_per_second": 4.743,
+    "train_steps_per_second": 0.009
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8824443293630243,
-  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM_Classification_base_V10/checkpoint-105",
+  "best_metric": 0.8720870015535992,
+  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM_Classification_base_V10/checkpoint-61",
   "epoch": 6.885245901639344,
   "eval_steps": 500,
   "global_step": 105,
@@ -10,141 +10,211 @@
   "log_history": [
     {
       "epoch": 0.6557377049180327,
-      "grad_norm": 5.335843086242676,
+      "grad_norm": 3.5814733505249023,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.8134,
+      "loss": 0.8498,
       "step": 10
     },
     {
       "epoch": 0.9836065573770492,
-      "eval_accuracy": 0.8430864836872087,
-      "eval_auc": 0.9250639596370771,
-      "eval_loss": 0.419461190700531,
-      "eval_runtime": 205.2459,
-      "eval_samples_per_second": 9.408,
+      "step": 15,
+      "train_accuracy": 0.8171125759338245,
+      "train_auc": 0.8967571163903928,
+      "train_loss": 0.47165462374687195,
+      "train_runtime": 694.4657,
+      "train_samples_per_second": 11.141,
+      "train_steps_per_second": 0.088
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "eval_accuracy": 0.8327291558777835,
+      "eval_auc": 0.9183935491091096,
+      "eval_loss": 0.4327956438064575,
+      "eval_runtime": 205.3728,
+      "eval_samples_per_second": 9.402,
       "eval_steps_per_second": 0.078,
       "step": 15
     },
     {
       "epoch": 1.3114754098360657,
-      "grad_norm": 2.8889973163604736,
+      "grad_norm": 2.3799431324005127,
       "learning_rate": 4.5212765957446815e-05,
-      "loss": 0.5101,
+      "loss": 0.5123,
       "step": 20
     },
     {
       "epoch": 1.9672131147540983,
-      "grad_norm": 1.6868778467178345,
+      "grad_norm": 2.067809581756592,
       "learning_rate": 3.9893617021276594e-05,
-      "loss": 0.3996,
+      "loss": 0.4085,
       "step": 30
     },
     {
       "epoch": 1.9672131147540983,
-      "eval_accuracy": 0.8570688762299327,
-      "eval_auc": 0.9437553079992403,
-      "eval_loss": 0.3682312071323395,
-      "eval_runtime": 204.3267,
-      "eval_samples_per_second": 9.451,
+      "step": 30,
+      "train_accuracy": 0.8517513248028952,
+      "train_auc": 0.9387841345112529,
+      "train_loss": 0.36320993304252625,
+      "train_runtime": 684.0745,
+      "train_samples_per_second": 11.31,
+      "train_steps_per_second": 0.089
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "eval_accuracy": 0.8518902123252201,
+      "eval_auc": 0.9423845489697825,
+      "eval_loss": 0.37112149596214294,
+      "eval_runtime": 205.1458,
+      "eval_samples_per_second": 9.413,
       "eval_steps_per_second": 0.078,
       "step": 30
     },
     {
       "epoch": 2.6229508196721314,
-      "grad_norm": 1.9163786172866821,
+      "grad_norm": 2.11800479888916,
       "learning_rate": 3.4574468085106386e-05,
-      "loss": 0.3693,
+      "loss": 0.3717,
       "step": 40
     },
     {
       "epoch": 2.9508196721311473,
-      "eval_accuracy": 0.877783531848783,
-      "eval_auc": 0.9529726135197429,
-      "eval_loss": 0.324381947517395,
-      "eval_runtime": 203.8679,
-      "eval_samples_per_second": 9.472,
-      "eval_steps_per_second": 0.078,
+      "step": 45,
+      "train_accuracy": 0.864029985782603,
+      "train_auc": 0.9527023779434378,
+      "train_loss": 0.3227292597293854,
+      "train_runtime": 681.4027,
+      "train_samples_per_second": 11.355,
+      "train_steps_per_second": 0.09
+    },
+    {
+      "epoch": 2.9508196721311473,
+      "eval_accuracy": 0.8653547384774728,
+      "eval_auc": 0.9487342167219569,
+      "eval_loss": 0.3345213234424591,
+      "eval_runtime": 203.71,
+      "eval_samples_per_second": 9.479,
+      "eval_steps_per_second": 0.079,
       "step": 45
     },
     {
       "epoch": 3.278688524590164,
-      "grad_norm": 2.268683433532715,
+      "grad_norm": 4.481107711791992,
       "learning_rate": 2.925531914893617e-05,
-      "loss": 0.3326,
+      "loss": 0.3301,
       "step": 50
     },
     {
       "epoch": 3.9344262295081966,
-      "grad_norm": 2.837725877761841,
+      "grad_norm": 2.703953981399536,
       "learning_rate": 2.393617021276596e-05,
-      "loss": 0.3147,
+      "loss": 0.3196,
       "step": 60
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8793371310201968,
-      "eval_auc": 0.9557452970649672,
-      "eval_loss": 0.31093424558639526,
-      "eval_runtime": 204.9066,
-      "eval_samples_per_second": 9.424,
+      "step": 61,
+      "train_accuracy": 0.8733359183145922,
+      "train_auc": 0.9563002964315263,
+      "train_loss": 0.30748647451400757,
+      "train_runtime": 678.9663,
+      "train_samples_per_second": 11.395,
+      "train_steps_per_second": 0.09
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8720870015535992,
+      "eval_auc": 0.9534343762433686,
+      "eval_loss": 0.3244781792163849,
+      "eval_runtime": 204.3133,
+      "eval_samples_per_second": 9.451,
       "eval_steps_per_second": 0.078,
       "step": 61
     },
     {
       "epoch": 4.590163934426229,
-      "grad_norm": 1.7413060665130615,
+      "grad_norm": 1.967414379119873,
       "learning_rate": 1.8617021276595745e-05,
-      "loss": 0.3166,
+      "loss": 0.3279,
       "step": 70
     },
     {
       "epoch": 4.983606557377049,
-      "eval_accuracy": 0.8803728638011393,
-      "eval_auc": 0.9561359266346766,
-      "eval_loss": 0.3134761154651642,
-      "eval_runtime": 204.4948,
-      "eval_samples_per_second": 9.443,
-      "eval_steps_per_second": 0.078,
+      "step": 76,
+      "train_accuracy": 0.8813493602171384,
+      "train_auc": 0.9603495718478623,
+      "train_loss": 0.2974039912223816,
+      "train_runtime": 680.7385,
+      "train_samples_per_second": 11.366,
+      "train_steps_per_second": 0.09
+    },
+    {
+      "epoch": 4.983606557377049,
+      "eval_accuracy": 0.8694976696012429,
+      "eval_auc": 0.953884522107936,
+      "eval_loss": 0.3240814805030823,
+      "eval_runtime": 203.7726,
+      "eval_samples_per_second": 9.476,
+      "eval_steps_per_second": 0.079,
       "step": 76
     },
     {
       "epoch": 5.245901639344262,
-      "grad_norm": 1.7945168018341064,
+      "grad_norm": 1.8759346008300781,
       "learning_rate": 1.3297872340425532e-05,
-      "loss": 0.2921,
+      "loss": 0.3042,
       "step": 80
     },
     {
       "epoch": 5.901639344262295,
-      "grad_norm": 1.865674376487732,
+      "grad_norm": 2.394904136657715,
       "learning_rate": 7.97872340425532e-06,
-      "loss": 0.2918,
+      "loss": 0.2926,
       "step": 90
     },
     {
       "epoch": 5.967213114754099,
-      "eval_accuracy": 0.8746763335059554,
-      "eval_auc": 0.9568584001596542,
-      "eval_loss": 0.3116800785064697,
-      "eval_runtime": 205.4646,
-      "eval_samples_per_second": 9.398,
+      "step": 91,
+      "train_accuracy": 0.8858730774201887,
+      "train_auc": 0.9646418504220758,
+      "train_loss": 0.28083333373069763,
+      "train_runtime": 681.28,
+      "train_samples_per_second": 11.357,
+      "train_steps_per_second": 0.09
+    },
+    {
+      "epoch": 5.967213114754099,
+      "eval_accuracy": 0.8674262040393579,
+      "eval_auc": 0.9548863198183253,
+      "eval_loss": 0.3211652338504791,
+      "eval_runtime": 205.0822,
+      "eval_samples_per_second": 9.416,
       "eval_steps_per_second": 0.078,
       "step": 91
     },
     {
       "epoch": 6.557377049180328,
-      "grad_norm": 4.410208225250244,
+      "grad_norm": 1.7362953424453735,
       "learning_rate": 2.6595744680851065e-06,
-      "loss": 0.2849,
+      "loss": 0.2821,
       "step": 100
     },
     {
       "epoch": 6.885245901639344,
-      "eval_accuracy": 0.8824443293630243,
-      "eval_auc": 0.957557852599524,
-      "eval_loss": 0.30561181902885437,
-      "eval_runtime": 203.0816,
-      "eval_samples_per_second": 9.508,
+      "step": 105,
+      "train_accuracy": 0.8937572702597906,
+      "train_auc": 0.9682688664846578,
+      "train_loss": 0.26618629693984985,
+      "train_runtime": 676.8276,
+      "train_samples_per_second": 11.431,
+      "train_steps_per_second": 0.09
+    },
+    {
+      "epoch": 6.885245901639344,
+      "eval_accuracy": 0.8705334023821854,
+      "eval_auc": 0.9549996078144867,
+      "eval_loss": 0.31725937128067017,
+      "eval_runtime": 203.7289,
+      "eval_samples_per_second": 9.478,
       "eval_steps_per_second": 0.079,
       "step": 105
     },
@@ -152,10 +222,10 @@
       "epoch": 6.885245901639344,
       "step": 105,
       "total_flos": 4.1785312376666235e+18,
-      "train_loss": 0.38780884515671504,
-      "train_runtime": 6680.4521,
-      "train_samples_per_second": 8.107,
-      "train_steps_per_second": 0.016
+      "train_loss": 0.3947783969697498,
+      "train_runtime": 11418.3292,
+      "train_samples_per_second": 4.743,
+      "train_steps_per_second": 0.009
     }
   ],
   "logging_steps": 10,