Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 7.4567966957568e+16,
-    "train_loss": 0.5813349982102712,
-    "train_runtime": 56.867,
-    "train_samples_per_second": 52.755,
-    "train_steps_per_second": 0.422
+    "epoch": 7.0,
+    "total_flos": 1.73991922900992e+17,
+    "train_loss": 0.5463650396892002,
+    "train_runtime": 159.2288,
+    "train_samples_per_second": 43.962,
+    "train_steps_per_second": 0.352
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cac8a5b27c54601914a1d7849e75bfe5c56ee04933b82f2889ea3cb58ba4c8d7
+oid sha256:8fbc5b474b36ce3cb06dc5975daf85b403ad30f6748e73184e159ea26e78d8db
 size 110394865

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 7.4567966957568e+16,
-    "train_loss": 0.5813349982102712,
-    "train_runtime": 56.867,
-    "train_samples_per_second": 52.755,
-    "train_steps_per_second": 0.422
+    "epoch": 7.0,
+    "total_flos": 1.73991922900992e+17,
+    "train_loss": 0.5463650396892002,
+    "train_runtime": 159.2288,
+    "train_samples_per_second": 43.962,
+    "train_steps_per_second": 0.352
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.7375,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224/checkpoint-24",
-  "epoch": 3.0,
+  "best_metric": 0.7583333333333333,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224/checkpoint-48",
+  "epoch": 7.0,
   "eval_steps": 500,
-  "global_step": 24,
+  "global_step": 56,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,66 +11,120 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.7333333333333333,
-      "eval_loss": 0.5740450024604797,
-      "eval_runtime": 1.7884,
-      "eval_samples_per_second": 134.201,
-      "eval_steps_per_second": 4.473,
+      "eval_loss": 0.58155757188797,
+      "eval_runtime": 1.8232,
+      "eval_samples_per_second": 131.637,
+      "eval_steps_per_second": 4.388,
       "step": 8
     },
     {
       "epoch": 1.25,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.6033,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6478,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.7333333333333333,
-      "eval_loss": 0.5640280246734619,
-      "eval_runtime": 1.8333,
-      "eval_samples_per_second": 130.909,
-      "eval_steps_per_second": 4.364,
+      "eval_loss": 0.5633648037910461,
+      "eval_runtime": 1.8405,
+      "eval_samples_per_second": 130.396,
+      "eval_steps_per_second": 4.347,
       "step": 16
     },
     {
       "epoch": 2.5,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.5751,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5746,
       "step": 20
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.7375,
-      "eval_loss": 0.5567867159843445,
-      "eval_runtime": 1.9734,
-      "eval_samples_per_second": 121.615,
-      "eval_steps_per_second": 4.054,
+      "eval_loss": 0.5526280999183655,
+      "eval_runtime": 1.8724,
+      "eval_samples_per_second": 128.178,
+      "eval_steps_per_second": 4.273,
       "step": 24
     },
     {
-      "epoch": 3.0,
-      "step": 24,
-      "total_flos": 7.4567966957568e+16,
-      "train_loss": 0.5813349982102712,
-      "train_runtime": 56.867,
-      "train_samples_per_second": 52.755,
-      "train_steps_per_second": 0.422
+      "epoch": 3.75,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 0.5414,
+      "step": 30
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.73,
-      "eval_loss": 0.5407183766365051,
-      "eval_runtime": 70.3286,
-      "eval_samples_per_second": 4.266,
-      "eval_steps_per_second": 0.142,
-      "step": 24
+      "epoch": 4.0,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_loss": 0.6044233441352844,
+      "eval_runtime": 3.3038,
+      "eval_samples_per_second": 72.644,
+      "eval_steps_per_second": 2.421,
+      "step": 32
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.5159,
+      "step": 40
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7541666666666667,
+      "eval_loss": 0.5310245156288147,
+      "eval_runtime": 2.8227,
+      "eval_samples_per_second": 85.024,
+      "eval_steps_per_second": 2.834,
+      "step": 40
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7583333333333333,
+      "eval_loss": 0.5481430888175964,
+      "eval_runtime": 3.9583,
+      "eval_samples_per_second": 60.632,
+      "eval_steps_per_second": 2.021,
+      "step": 48
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 6e-06,
+      "loss": 0.4901,
+      "step": 50
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7583333333333333,
+      "eval_loss": 0.5410298109054565,
+      "eval_runtime": 2.4531,
+      "eval_samples_per_second": 97.834,
+      "eval_steps_per_second": 3.261,
+      "step": 56
+    },
+    {
+      "epoch": 7.0,
+      "step": 56,
+      "total_flos": 1.73991922900992e+17,
+      "train_loss": 0.5463650396892002,
+      "train_runtime": 159.2288,
+      "train_samples_per_second": 43.962,
+      "train_steps_per_second": 0.352
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7366666666666667,
+      "eval_loss": 0.5300371050834656,
+      "eval_runtime": 4.2651,
+      "eval_samples_per_second": 70.339,
+      "eval_steps_per_second": 2.345,
+      "step": 56
     }
   ],
   "logging_steps": 10,
-  "max_steps": 24,
-  "num_train_epochs": 3,
+  "max_steps": 56,
+  "num_train_epochs": 7,
   "save_steps": 500,
-  "total_flos": 7.4567966957568e+16,
+  "total_flos": 1.73991922900992e+17,
   "trial_name": null,
   "trial_params": null
 }