Training in progress, epoch 6

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.7880986937590712,
-    "eval_loss": 0.5690365433692932,
-    "eval_runtime": 34.5814,
-    "eval_samples_per_second": 39.848,
-    "eval_steps_per_second": 1.272,
-    "total_flos": 9.244220036417372e+17,
-    "train_loss": 0.4883061381140116,
-    "train_runtime": 824.2717,
-    "train_samples_per_second": 45.116,
-    "train_steps_per_second": 0.353
+    "epoch": 5.0,
+    "eval_accuracy": 0.8084179970972424,
+    "eval_loss": 0.5213053226470947,
+    "eval_runtime": 34.1734,
+    "eval_samples_per_second": 40.324,
+    "eval_steps_per_second": 1.288,
+    "total_flos": 1.5407033394028954e+18,
+    "train_loss": 0.25417212948356704,
+    "train_runtime": 754.4258,
+    "train_samples_per_second": 82.155,
+    "train_steps_per_second": 0.643
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.7880986937590712,
-    "eval_loss": 0.5690365433692932,
-    "eval_runtime": 34.5814,
-    "eval_samples_per_second": 39.848,
-    "eval_steps_per_second": 1.272
+    "epoch": 5.0,
+    "eval_accuracy": 0.8084179970972424,
+    "eval_loss": 0.5213053226470947,
+    "eval_runtime": 34.1734,
+    "eval_samples_per_second": 40.324,
+    "eval_steps_per_second": 1.288
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:904ab153a33d8ef1bfe0e4766d15493636392f11a420eec77d91ab1e1343fa78
+oid sha256:319ff3acee313dc8a0ac65b91528f4931298f23767e217e3f39b64886437b1bb
 size 110352060

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 9.244220036417372e+17,
-    "train_loss": 0.4883061381140116,
-    "train_runtime": 824.2717,
-    "train_samples_per_second": 45.116,
-    "train_steps_per_second": 0.353
+    "epoch": 5.0,
+    "total_flos": 1.5407033394028954e+18,
+    "train_loss": 0.25417212948356704,
+    "train_runtime": 754.4258,
+    "train_samples_per_second": 82.155,
+    "train_steps_per_second": 0.643
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.7880986937590712,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-291",
-  "epoch": 3.0,
+  "best_metric": 0.8084179970972424,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-485",
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 291,
+  "global_step": 485,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -210,20 +210,152 @@
       "step": 291
     },
     {
-      "epoch": 3.0,
-      "step": 291,
-      "total_flos": 9.244220036417372e+17,
-      "train_loss": 0.4883061381140116,
-      "train_runtime": 824.2717,
-      "train_samples_per_second": 45.116,
-      "train_steps_per_second": 0.353
+      "epoch": 3.09,
+      "learning_rate": 2.1215596330275232e-05,
+      "loss": 0.6839,
+      "step": 300
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 2.006880733944954e-05,
+      "loss": 0.6814,
+      "step": 310
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 1.8922018348623855e-05,
+      "loss": 0.6358,
+      "step": 320
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 1.7775229357798164e-05,
+      "loss": 0.6343,
+      "step": 330
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.662844036697248e-05,
+      "loss": 0.6434,
+      "step": 340
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 1.548165137614679e-05,
+      "loss": 0.6701,
+      "step": 350
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 1.4334862385321102e-05,
+      "loss": 0.6605,
+      "step": 360
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 1.3188073394495412e-05,
+      "loss": 0.6269,
+      "step": 370
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.2041284403669725e-05,
+      "loss": 0.6522,
+      "step": 380
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7968069666182874,
+      "eval_loss": 0.5324537754058838,
+      "eval_runtime": 34.108,
+      "eval_samples_per_second": 40.401,
+      "eval_steps_per_second": 1.29,
+      "step": 388
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 1.0894495412844036e-05,
+      "loss": 0.6545,
+      "step": 390
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 9.74770642201835e-06,
+      "loss": 0.6333,
+      "step": 400
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 8.600917431192661e-06,
+      "loss": 0.5846,
+      "step": 410
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 7.454128440366973e-06,
+      "loss": 0.6449,
+      "step": 420
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 6.3073394495412846e-06,
+      "loss": 0.586,
+      "step": 430
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 5.160550458715597e-06,
+      "loss": 0.6197,
+      "step": 440
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 4.013761467889909e-06,
+      "loss": 0.5752,
+      "step": 450
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 2.8669724770642206e-06,
+      "loss": 0.6203,
+      "step": 460
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 1.7201834862385322e-06,
+      "loss": 0.6595,
+      "step": 470
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 5.733944954128441e-07,
+      "loss": 0.6189,
+      "step": 480
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8084179970972424,
+      "eval_loss": 0.5213053226470947,
+      "eval_runtime": 34.2264,
+      "eval_samples_per_second": 40.261,
+      "eval_steps_per_second": 1.286,
+      "step": 485
+    },
+    {
+      "epoch": 5.0,
+      "step": 485,
+      "total_flos": 1.5407033394028954e+18,
+      "train_loss": 0.25417212948356704,
+      "train_runtime": 754.4258,
+      "train_samples_per_second": 82.155,
+      "train_steps_per_second": 0.643
     }
   ],
   "logging_steps": 10,
-  "max_steps": 291,
-  "num_train_epochs": 3,
+  "max_steps": 485,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 9.244220036417372e+17,
+  "total_flos": 1.5407033394028954e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b731c7fb02690e62840d5fde0f3f479264c690e9befb062d41e3f9d1b5f143a1
+oid sha256:f699c4d6e1218ced6a2b6d295a8eaf205ad55fdd9d20805536e894a146aad539
 size 4219