Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "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
+    "train_loss": 0.5857737362384796,
+    "train_runtime": 257.742,
+    "train_samples_per_second": 27.159,
+    "train_steps_per_second": 0.217
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3767eccbee6326874ee4ee31faf86b112c49f69226fb388dad681c5c9a0ad176
+oid sha256:12b05e450262fc68a5119dd5a3fb702fe0f750f289d47e5fd3597756068b0481
 size 110394865

train_results.json

@@ -1,8 +1,8 @@
 {
     "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
+    "train_loss": 0.5857737362384796,
+    "train_runtime": 257.742,
+    "train_samples_per_second": 27.159,
+    "train_steps_per_second": 0.217
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7583333333333333,
+  "best_metric": 0.7416666666666667,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224/checkpoint-48",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -11,112 +11,136 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.7333333333333333,
-      "eval_loss": 0.58155757188797,
-      "eval_runtime": 1.8232,
-      "eval_samples_per_second": 131.637,
-      "eval_steps_per_second": 4.388,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5834997296333313,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 33.1193,
+      "eval_samples_per_second": 7.247,
+      "eval_steps_per_second": 0.242,
       "step": 8
     },
     {
       "epoch": 1.25,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.6478,
+      "loss": 0.6395,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.7333333333333333,
-      "eval_loss": 0.5633648037910461,
-      "eval_runtime": 1.8405,
-      "eval_samples_per_second": 130.396,
-      "eval_steps_per_second": 4.347,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5735692381858826,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 2.0727,
+      "eval_samples_per_second": 115.793,
+      "eval_steps_per_second": 3.86,
       "step": 16
     },
     {
       "epoch": 2.5,
       "learning_rate": 3.6e-05,
-      "loss": 0.5746,
+      "loss": 0.5789,
       "step": 20
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7375,
-      "eval_loss": 0.5526280999183655,
-      "eval_runtime": 1.8724,
-      "eval_samples_per_second": 128.178,
-      "eval_steps_per_second": 4.273,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5942796468734741,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 2.0683,
+      "eval_samples_per_second": 116.035,
+      "eval_steps_per_second": 3.868,
       "step": 24
     },
     {
       "epoch": 3.75,
       "learning_rate": 2.6000000000000002e-05,
-      "loss": 0.5414,
+      "loss": 0.5887,
       "step": 30
     },
     {
       "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,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5613328218460083,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 2.0585,
+      "eval_samples_per_second": 116.592,
+      "eval_steps_per_second": 3.886,
       "step": 32
     },
     {
       "epoch": 5.0,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.5159,
+      "loss": 0.5791,
       "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,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5557453036308289,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 2.0897,
+      "eval_samples_per_second": 114.847,
+      "eval_steps_per_second": 3.828,
       "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,
+      "eval_accuracy": 0.7416666666666667,
+      "eval_f1_score": 0.6396720831503441,
+      "eval_loss": 0.5535091161727905,
+      "eval_precision": 0.8089635854341737,
+      "eval_recall": 0.7416666666666667,
+      "eval_runtime": 1.9813,
+      "eval_samples_per_second": 121.134,
+      "eval_steps_per_second": 4.038,
       "step": 48
     },
     {
       "epoch": 6.25,
       "learning_rate": 6e-06,
-      "loss": 0.4901,
+      "loss": 0.5717,
       "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,
+      "eval_accuracy": 0.7333333333333333,
+      "eval_f1_score": 0.6205128205128204,
+      "eval_loss": 0.5456091165542603,
+      "eval_precision": 0.5377777777777778,
+      "eval_recall": 0.7333333333333333,
+      "eval_runtime": 2.0655,
+      "eval_samples_per_second": 116.197,
+      "eval_steps_per_second": 3.873,
       "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
+      "train_loss": 0.5857737362384796,
+      "train_runtime": 257.742,
+      "train_samples_per_second": 27.159,
+      "train_steps_per_second": 0.217
     },
     {
       "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,
+      "eval_f1_score": 0.6314646594113398,
+      "eval_loss": 0.5486770868301392,
+      "eval_precision": 0.8064765100671142,
+      "eval_recall": 0.7366666666666667,
+      "eval_runtime": 109.0657,
+      "eval_samples_per_second": 2.751,
+      "eval_steps_per_second": 0.092,
       "step": 56
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5aff4bdc67ae077b0bf7d175151eed746c480633a9d9571a02c41d85a9abb5f2
+oid sha256:fb7e8116dbe569fb96ff854b905cb0bc3ae5795528e7afede2c8328a68bddc7d
 size 4091