Training in progress, epoch 1

all_results.json

@@ -6,8 +6,8 @@
     "eval_samples_per_second": 96.41,
     "eval_steps_per_second": 4.821,
     "total_flos": 7.078985663171789e+16,
-    "train_loss": 0.5234465897083282,
-    "train_runtime": 62.5923,
-    "train_samples_per_second": 45.501,
-    "train_steps_per_second": 0.383
+    "train_loss": 0.5273787677288055,
+    "train_runtime": 64.2353,
+    "train_samples_per_second": 44.337,
+    "train_steps_per_second": 0.374
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abd1dd65ff1a6e1e819c0f0e363d8601d94c85066720fd0ace6ddf6887fed709
+oid sha256:f01aab063f428203a00702c86614c37542bb7e7421a739c0ed52034fce0411c0
 size 110394865

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 8.0,
     "total_flos": 7.078985663171789e+16,
-    "train_loss": 0.5234465897083282,
-    "train_runtime": 62.5923,
-    "train_samples_per_second": 45.501,
-    "train_steps_per_second": 0.383
+    "train_loss": 0.5273787677288055,
+    "train_runtime": 64.2353,
+    "train_samples_per_second": 44.337,
+    "train_steps_per_second": 0.374
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.7647058823529411,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetunedo/checkpoint-6",
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetunedo/checkpoint-3",
   "epoch": 8.0,
   "eval_steps": 500,
   "global_step": 24,
@@ -10,96 +10,96 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_loss": 0.627018928527832,
-      "eval_roc_auc": 0.5588235294117647,
-      "eval_runtime": 0.3391,
-      "eval_samples_per_second": 117.954,
-      "eval_steps_per_second": 5.898,
+      "eval_loss": 0.543735682964325,
+      "eval_roc_auc": 0.7647058823529411,
+      "eval_runtime": 0.2816,
+      "eval_samples_per_second": 142.031,
+      "eval_steps_per_second": 7.102,
       "step": 3
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.5486534833908081,
+      "eval_loss": 0.5345174074172974,
       "eval_roc_auc": 0.7647058823529411,
-      "eval_runtime": 0.2922,
-      "eval_samples_per_second": 136.875,
-      "eval_steps_per_second": 6.844,
+      "eval_runtime": 0.2839,
+      "eval_samples_per_second": 140.898,
+      "eval_steps_per_second": 7.045,
       "step": 6
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.5324329733848572,
-      "eval_roc_auc": 0.6854219948849105,
-      "eval_runtime": 0.3456,
-      "eval_samples_per_second": 115.744,
-      "eval_steps_per_second": 5.787,
+      "eval_loss": 0.5251768827438354,
+      "eval_roc_auc": 0.7647058823529411,
+      "eval_runtime": 0.2927,
+      "eval_samples_per_second": 136.645,
+      "eval_steps_per_second": 6.832,
       "step": 9
     },
     {
       "epoch": 3.33,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.6124,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.5245,
       "step": 10
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.5400528907775879,
-      "eval_roc_auc": 0.6854219948849105,
-      "eval_runtime": 0.3025,
-      "eval_samples_per_second": 132.249,
-      "eval_steps_per_second": 6.612,
+      "eval_loss": 0.5188445448875427,
+      "eval_roc_auc": 0.7647058823529411,
+      "eval_runtime": 0.2961,
+      "eval_samples_per_second": 135.085,
+      "eval_steps_per_second": 6.754,
       "step": 12
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.5094296932220459,
-      "eval_roc_auc": 0.6994884910485933,
-      "eval_runtime": 0.2822,
-      "eval_samples_per_second": 141.73,
-      "eval_steps_per_second": 7.086,
+      "eval_loss": 0.5151057839393616,
+      "eval_roc_auc": 0.7647058823529411,
+      "eval_runtime": 0.2892,
+      "eval_samples_per_second": 138.29,
+      "eval_steps_per_second": 6.914,
       "step": 15
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.4883674085140228,
-      "eval_roc_auc": 0.7212276214833759,
-      "eval_runtime": 0.2802,
-      "eval_samples_per_second": 142.735,
-      "eval_steps_per_second": 7.137,
+      "eval_loss": 0.5121785402297974,
+      "eval_roc_auc": 0.7429667519181585,
+      "eval_runtime": 0.2877,
+      "eval_samples_per_second": 139.044,
+      "eval_steps_per_second": 6.952,
       "step": 18
     },
     {
       "epoch": 6.67,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.4581,
+      "learning_rate": 9.523809523809525e-07,
+      "loss": 0.5211,
       "step": 20
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.4820639193058014,
-      "eval_roc_auc": 0.6777493606138107,
-      "eval_runtime": 0.2832,
-      "eval_samples_per_second": 141.223,
-      "eval_steps_per_second": 7.061,
+      "eval_loss": 0.5102273225784302,
+      "eval_roc_auc": 0.7429667519181585,
+      "eval_runtime": 0.3519,
+      "eval_samples_per_second": 113.673,
+      "eval_steps_per_second": 5.684,
       "step": 21
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.48128217458724976,
-      "eval_roc_auc": 0.6854219948849105,
-      "eval_runtime": 0.4605,
-      "eval_samples_per_second": 86.857,
-      "eval_steps_per_second": 4.343,
+      "eval_loss": 0.509425938129425,
+      "eval_roc_auc": 0.7429667519181585,
+      "eval_runtime": 0.3558,
+      "eval_samples_per_second": 112.422,
+      "eval_steps_per_second": 5.621,
       "step": 24
     },
     {
       "epoch": 8.0,
       "step": 24,
       "total_flos": 7.078985663171789e+16,
-      "train_loss": 0.5234465897083282,
-      "train_runtime": 62.5923,
-      "train_samples_per_second": 45.501,
-      "train_steps_per_second": 0.383
+      "train_loss": 0.5273787677288055,
+      "train_runtime": 64.2353,
+      "train_samples_per_second": 44.337,
+      "train_steps_per_second": 0.374
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0120c3097b608f3d5517eb4e150cfcc14607bafaa638f00789df51dae77c839
+oid sha256:a43206ac8f49b4691c8f2b5ac864e908ef80a0c86db3465b4ecd328f4f584c34
 size 4091