Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 9.68421052631579,
-    "eval_accuracy": 0.9286657859973579,
-    "eval_loss": 0.19915194809436798,
-    "eval_runtime": 3.1234,
-    "eval_samples_per_second": 242.36,
-    "eval_steps_per_second": 7.684,
+    "eval_accuracy": 0.9352708058124174,
+    "eval_loss": 0.1843554824590683,
+    "eval_runtime": 3.1532,
+    "eval_samples_per_second": 240.077,
+    "eval_steps_per_second": 7.611,
     "total_flos": 7.291573574754632e+17,
-    "train_loss": 0.36391739119654115,
-    "train_runtime": 309.8219,
-    "train_samples_per_second": 97.734,
-    "train_steps_per_second": 0.742
+    "train_loss": 0.36177816701971965,
+    "train_runtime": 310.2655,
+    "train_samples_per_second": 97.594,
+    "train_steps_per_second": 0.741
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.68421052631579,
-    "eval_accuracy": 0.9286657859973579,
-    "eval_loss": 0.19915194809436798,
-    "eval_runtime": 3.1234,
-    "eval_samples_per_second": 242.36,
-    "eval_steps_per_second": 7.684
+    "eval_accuracy": 0.9352708058124174,
+    "eval_loss": 0.1843554824590683,
+    "eval_runtime": 3.1532,
+    "eval_samples_per_second": 240.077,
+    "eval_steps_per_second": 7.611
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f93be9cc37b268b0eaf6cd2897c60f94fdd215a16d649e24ec1cfa1a5d10ec0
+oid sha256:9b80417fbab649d6952234f47689d844e63a0c4201df74afdfcb488c360b85e7
 size 110355136

runs/Jun10_05-30-11_e6d590d50f6e/events.out.tfevents.1717997741.e6d590d50f6e.461.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:895742b41821b52486dd8476afe46a30078a32b0547b591caa9cb2fc4789694b
+size 411

runs/Jun10_05-35-50_e6d590d50f6e/events.out.tfevents.1717997751.e6d590d50f6e.461.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6f8f70de6a3163db263348d0591e69c1b2549d33da6a43c69bff695f8ce939b
+size 6066

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.68421052631579,
     "total_flos": 7.291573574754632e+17,
-    "train_loss": 0.36391739119654115,
-    "train_runtime": 309.8219,
-    "train_samples_per_second": 97.734,
-    "train_steps_per_second": 0.742
+    "train_loss": 0.36177816701971965,
+    "train_runtime": 310.2655,
+    "train_samples_per_second": 97.594,
+    "train_steps_per_second": 0.741
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9286657859973579,
+  "best_metric": 0.9352708058124174,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224-finalterm/checkpoint-166",
   "epoch": 9.68421052631579,
   "eval_steps": 500,
@@ -10,263 +10,263 @@
   "log_history": [
     {
       "epoch": 0.42105263157894735,
-      "grad_norm": 4.981729030609131,
+      "grad_norm": 6.019707679748535,
       "learning_rate": 2.173913043478261e-05,
-      "loss": 1.7762,
+      "loss": 1.7902,
       "step": 10
     },
     {
       "epoch": 0.8421052631578947,
-      "grad_norm": 6.3064351081848145,
+      "grad_norm": 5.092822551727295,
       "learning_rate": 4.347826086956522e-05,
-      "loss": 1.2941,
+      "loss": 1.2727,
       "step": 20
     },
     {
       "epoch": 0.968421052631579,
-      "eval_accuracy": 0.8071334214002642,
-      "eval_loss": 0.6089140176773071,
-      "eval_runtime": 3.1553,
-      "eval_samples_per_second": 239.917,
-      "eval_steps_per_second": 7.606,
+      "eval_accuracy": 0.8335535006605019,
+      "eval_loss": 0.5535598993301392,
+      "eval_runtime": 3.1512,
+      "eval_samples_per_second": 240.223,
+      "eval_steps_per_second": 7.616,
       "step": 23
     },
     {
       "epoch": 1.263157894736842,
-      "grad_norm": 7.853344440460205,
+      "grad_norm": 6.4061150550842285,
       "learning_rate": 4.830917874396135e-05,
-      "loss": 0.631,
+      "loss": 0.5623,
       "step": 30
     },
     {
       "epoch": 1.6842105263157894,
-      "grad_norm": 6.575767993927002,
+      "grad_norm": 4.992868900299072,
       "learning_rate": 4.589371980676328e-05,
-      "loss": 0.4373,
+      "loss": 0.3845,
       "step": 40
     },
     {
       "epoch": 1.9789473684210526,
-      "eval_accuracy": 0.8956406869220608,
-      "eval_loss": 0.2909472584724426,
-      "eval_runtime": 3.1122,
-      "eval_samples_per_second": 243.24,
-      "eval_steps_per_second": 7.712,
+      "eval_accuracy": 0.9114927344782034,
+      "eval_loss": 0.23858527839183807,
+      "eval_runtime": 3.1195,
+      "eval_samples_per_second": 242.671,
+      "eval_steps_per_second": 7.694,
       "step": 47
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 6.464214324951172,
+      "grad_norm": 5.19857120513916,
       "learning_rate": 4.347826086956522e-05,
-      "loss": 0.3887,
+      "loss": 0.3797,
       "step": 50
     },
     {
       "epoch": 2.526315789473684,
-      "grad_norm": 4.2318806648254395,
+      "grad_norm": 6.890084743499756,
       "learning_rate": 4.106280193236715e-05,
-      "loss": 0.3111,
+      "loss": 0.3025,
       "step": 60
     },
     {
       "epoch": 2.9473684210526314,
-      "grad_norm": 5.7080464363098145,
+      "grad_norm": 4.1722092628479,
       "learning_rate": 3.864734299516908e-05,
-      "loss": 0.2955,
+      "loss": 0.2725,
       "step": 70
     },
     {
       "epoch": 2.9894736842105263,
-      "eval_accuracy": 0.9180977542932629,
-      "eval_loss": 0.23408588767051697,
-      "eval_runtime": 3.1384,
-      "eval_samples_per_second": 241.208,
-      "eval_steps_per_second": 7.647,
+      "eval_accuracy": 0.9233817701453104,
+      "eval_loss": 0.21346069872379303,
+      "eval_runtime": 3.2718,
+      "eval_samples_per_second": 231.37,
+      "eval_steps_per_second": 7.335,
       "step": 71
     },
     {
       "epoch": 3.3684210526315788,
-      "grad_norm": 5.401839256286621,
+      "grad_norm": 3.6963953971862793,
       "learning_rate": 3.6231884057971014e-05,
-      "loss": 0.2439,
+      "loss": 0.2404,
       "step": 80
     },
     {
       "epoch": 3.7894736842105265,
-      "grad_norm": 7.044283866882324,
+      "grad_norm": 4.915622711181641,
       "learning_rate": 3.381642512077295e-05,
-      "loss": 0.2618,
+      "loss": 0.2442,
       "step": 90
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9247027741083224,
-      "eval_loss": 0.21280620992183685,
-      "eval_runtime": 3.1094,
-      "eval_samples_per_second": 243.454,
-      "eval_steps_per_second": 7.718,
+      "eval_accuracy": 0.9075297225891678,
+      "eval_loss": 0.2290719598531723,
+      "eval_runtime": 3.1408,
+      "eval_samples_per_second": 241.02,
+      "eval_steps_per_second": 7.641,
       "step": 95
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 3.9932973384857178,
+      "grad_norm": 3.525477170944214,
       "learning_rate": 3.140096618357488e-05,
-      "loss": 0.2361,
+      "loss": 0.265,
       "step": 100
     },
     {
       "epoch": 4.631578947368421,
-      "grad_norm": 6.241832733154297,
+      "grad_norm": 4.751044750213623,
       "learning_rate": 2.8985507246376814e-05,
-      "loss": 0.2464,
+      "loss": 0.2097,
       "step": 110
     },
     {
       "epoch": 4.968421052631579,
-      "eval_accuracy": 0.916776750330251,
-      "eval_loss": 0.22047297656536102,
-      "eval_runtime": 3.1366,
-      "eval_samples_per_second": 241.348,
-      "eval_steps_per_second": 7.652,
+      "eval_accuracy": 0.9207397622192867,
+      "eval_loss": 0.19642269611358643,
+      "eval_runtime": 3.1107,
+      "eval_samples_per_second": 243.355,
+      "eval_steps_per_second": 7.715,
       "step": 118
     },
     {
       "epoch": 5.052631578947368,
-      "grad_norm": 4.780943393707275,
+      "grad_norm": 6.279447078704834,
       "learning_rate": 2.6570048309178748e-05,
-      "loss": 0.2331,
+      "loss": 0.2359,
       "step": 120
     },
     {
       "epoch": 5.473684210526316,
-      "grad_norm": 6.229685306549072,
+      "grad_norm": 5.107997894287109,
       "learning_rate": 2.4154589371980676e-05,
-      "loss": 0.2137,
+      "loss": 0.2293,
       "step": 130
     },
     {
       "epoch": 5.894736842105263,
-      "grad_norm": 4.2907023429870605,
+      "grad_norm": 4.217101097106934,
       "learning_rate": 2.173913043478261e-05,
-      "loss": 0.1988,
+      "loss": 0.2237,
       "step": 140
     },
     {
       "epoch": 5.978947368421053,
-      "eval_accuracy": 0.9194187582562747,
-      "eval_loss": 0.2052578181028366,
-      "eval_runtime": 3.1312,
-      "eval_samples_per_second": 241.757,
-      "eval_steps_per_second": 7.665,
+      "eval_accuracy": 0.9286657859973579,
+      "eval_loss": 0.19202813506126404,
+      "eval_runtime": 3.1318,
+      "eval_samples_per_second": 241.714,
+      "eval_steps_per_second": 7.663,
       "step": 142
     },
     {
       "epoch": 6.315789473684211,
-      "grad_norm": 4.971545219421387,
+      "grad_norm": 4.676567077636719,
       "learning_rate": 1.932367149758454e-05,
-      "loss": 0.1824,
+      "loss": 0.1789,
       "step": 150
     },
     {
       "epoch": 6.7368421052631575,
-      "grad_norm": 5.901086330413818,
+      "grad_norm": 4.55054235458374,
       "learning_rate": 1.6908212560386476e-05,
-      "loss": 0.1954,
+      "loss": 0.2199,
       "step": 160
     },
     {
       "epoch": 6.989473684210527,
-      "eval_accuracy": 0.9286657859973579,
-      "eval_loss": 0.19915194809436798,
-      "eval_runtime": 3.1105,
-      "eval_samples_per_second": 243.373,
-      "eval_steps_per_second": 7.716,
+      "eval_accuracy": 0.9352708058124174,
+      "eval_loss": 0.1843554824590683,
+      "eval_runtime": 3.1189,
+      "eval_samples_per_second": 242.711,
+      "eval_steps_per_second": 7.695,
       "step": 166
     },
     {
       "epoch": 7.157894736842105,
-      "grad_norm": 5.259510040283203,
+      "grad_norm": 4.944462776184082,
       "learning_rate": 1.4492753623188407e-05,
-      "loss": 0.2037,
+      "loss": 0.2016,
       "step": 170
     },
     {
       "epoch": 7.578947368421053,
-      "grad_norm": 4.991972923278809,
+      "grad_norm": 4.52495813369751,
       "learning_rate": 1.2077294685990338e-05,
-      "loss": 0.1865,
+      "loss": 0.1954,
       "step": 180
     },
     {
       "epoch": 8.0,
-      "grad_norm": 5.169896602630615,
+      "grad_norm": 5.276759624481201,
       "learning_rate": 9.66183574879227e-06,
-      "loss": 0.1749,
+      "loss": 0.2209,
       "step": 190
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9260237780713342,
-      "eval_loss": 0.19742345809936523,
-      "eval_runtime": 3.1731,
-      "eval_samples_per_second": 238.569,
-      "eval_steps_per_second": 7.564,
+      "eval_accuracy": 0.9273447820343461,
+      "eval_loss": 0.18566328287124634,
+      "eval_runtime": 3.1379,
+      "eval_samples_per_second": 241.243,
+      "eval_steps_per_second": 7.648,
       "step": 190
     },
     {
       "epoch": 8.421052631578947,
-      "grad_norm": 3.162759780883789,
+      "grad_norm": 4.787791728973389,
       "learning_rate": 7.246376811594203e-06,
-      "loss": 0.1521,
+      "loss": 0.1914,
       "step": 200
     },
     {
       "epoch": 8.842105263157894,
-      "grad_norm": 8.834267616271973,
+      "grad_norm": 4.165464401245117,
       "learning_rate": 4.830917874396135e-06,
-      "loss": 0.1621,
+      "loss": 0.1717,
       "step": 210
     },
     {
       "epoch": 8.968421052631578,
-      "eval_accuracy": 0.9273447820343461,
-      "eval_loss": 0.19469071924686432,
-      "eval_runtime": 3.1137,
-      "eval_samples_per_second": 243.115,
-      "eval_steps_per_second": 7.708,
+      "eval_accuracy": 0.9313077939233818,
+      "eval_loss": 0.18422812223434448,
+      "eval_runtime": 3.1181,
+      "eval_samples_per_second": 242.779,
+      "eval_steps_per_second": 7.697,
       "step": 213
     },
     {
       "epoch": 9.263157894736842,
-      "grad_norm": 5.654167652130127,
+      "grad_norm": 2.85178279876709,
       "learning_rate": 2.4154589371980677e-06,
-      "loss": 0.1972,
+      "loss": 0.1532,
       "step": 220
     },
     {
       "epoch": 9.68421052631579,
-      "grad_norm": 3.4447648525238037,
+      "grad_norm": 4.912642478942871,
       "learning_rate": 0.0,
-      "loss": 0.1482,
+      "loss": 0.1754,
       "step": 230
     },
     {
       "epoch": 9.68421052631579,
-      "eval_accuracy": 0.9273447820343461,
-      "eval_loss": 0.1953066885471344,
-      "eval_runtime": 3.099,
-      "eval_samples_per_second": 244.276,
-      "eval_steps_per_second": 7.745,
+      "eval_accuracy": 0.9313077939233818,
+      "eval_loss": 0.18370747566223145,
+      "eval_runtime": 3.1098,
+      "eval_samples_per_second": 243.427,
+      "eval_steps_per_second": 7.718,
       "step": 230
     },
     {
       "epoch": 9.68421052631579,
       "step": 230,
       "total_flos": 7.291573574754632e+17,
-      "train_loss": 0.36391739119654115,
-      "train_runtime": 309.8219,
-      "train_samples_per_second": 97.734,
-      "train_steps_per_second": 0.742
+      "train_loss": 0.36177816701971965,
+      "train_runtime": 310.2655,
+      "train_samples_per_second": 97.594,
+      "train_steps_per_second": 0.741
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7564323cc580936ad71f47e4ccb2e4a4f25746d4048283673f5c4fbe17b8ea2e
+oid sha256:94ef88ce9caefe0cc757b7eb234413ab26e67b39e3200d8c43ac067a6246a1d2
 size 5176