Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.0,
     "total_flos": 1.73991922900992e+17,
-    "train_loss": 0.5857737362384796,
-    "train_runtime": 257.742,
-    "train_samples_per_second": 27.159,
-    "train_steps_per_second": 0.217
+    "train_loss": 0.5527358267988477,
+    "train_runtime": 138.7489,
+    "train_samples_per_second": 50.451,
+    "train_steps_per_second": 0.404
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b547a75b48f9d72a3e5efcc75583ec98f616f639143ccea478167058a79af3e
+oid sha256:b0a20659b3ad9192cfbe75a1ac3cab84dd5423b7ef52b01beafb000977e0e963
 size 110394865

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.0,
     "total_flos": 1.73991922900992e+17,
-    "train_loss": 0.5857737362384796,
-    "train_runtime": 257.742,
-    "train_samples_per_second": 27.159,
-    "train_steps_per_second": 0.217
+    "train_loss": 0.5527358267988477,
+    "train_runtime": 138.7489,
+    "train_samples_per_second": 50.451,
+    "train_steps_per_second": 0.404
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7416666666666667,
+  "best_metric": 0.7625,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224/checkpoint-48",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -12,135 +12,135 @@
       "epoch": 1.0,
       "eval_accuracy": 0.7333333333333333,
       "eval_f1_score": 0.6205128205128204,
-      "eval_loss": 0.5834997296333313,
+      "eval_loss": 0.5696213841438293,
       "eval_precision": 0.5377777777777778,
       "eval_recall": 0.7333333333333333,
-      "eval_runtime": 33.1193,
-      "eval_samples_per_second": 7.247,
-      "eval_steps_per_second": 0.242,
+      "eval_runtime": 1.8813,
+      "eval_samples_per_second": 127.574,
+      "eval_steps_per_second": 4.252,
       "step": 8
     },
     {
       "epoch": 1.25,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.6395,
+      "loss": 0.6683,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.7333333333333333,
       "eval_f1_score": 0.6205128205128204,
-      "eval_loss": 0.5735692381858826,
+      "eval_loss": 0.5635157227516174,
       "eval_precision": 0.5377777777777778,
       "eval_recall": 0.7333333333333333,
-      "eval_runtime": 2.0727,
-      "eval_samples_per_second": 115.793,
-      "eval_steps_per_second": 3.86,
+      "eval_runtime": 1.8341,
+      "eval_samples_per_second": 130.853,
+      "eval_steps_per_second": 4.362,
       "step": 16
     },
     {
       "epoch": 2.5,
       "learning_rate": 3.6e-05,
-      "loss": 0.5789,
+      "loss": 0.5797,
       "step": 20
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.7333333333333333,
       "eval_f1_score": 0.6205128205128204,
-      "eval_loss": 0.5942796468734741,
+      "eval_loss": 0.55841064453125,
       "eval_precision": 0.5377777777777778,
       "eval_recall": 0.7333333333333333,
-      "eval_runtime": 2.0683,
-      "eval_samples_per_second": 116.035,
-      "eval_steps_per_second": 3.868,
+      "eval_runtime": 2.0336,
+      "eval_samples_per_second": 118.017,
+      "eval_steps_per_second": 3.934,
       "step": 24
     },
     {
       "epoch": 3.75,
       "learning_rate": 2.6000000000000002e-05,
-      "loss": 0.5887,
+      "loss": 0.5547,
       "step": 30
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.7333333333333333,
       "eval_f1_score": 0.6205128205128204,
-      "eval_loss": 0.5613328218460083,
+      "eval_loss": 0.5731536746025085,
       "eval_precision": 0.5377777777777778,
       "eval_recall": 0.7333333333333333,
-      "eval_runtime": 2.0585,
-      "eval_samples_per_second": 116.592,
-      "eval_steps_per_second": 3.886,
+      "eval_runtime": 1.8352,
+      "eval_samples_per_second": 130.778,
+      "eval_steps_per_second": 4.359,
       "step": 32
     },
     {
       "epoch": 5.0,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.5791,
+      "loss": 0.5165,
       "step": 40
     },
     {
       "epoch": 5.0,
-      "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,
+      "eval_accuracy": 0.7583333333333333,
+      "eval_f1_score": 0.6959050899349406,
+      "eval_loss": 0.5416003465652466,
+      "eval_precision": 0.7485882848714707,
+      "eval_recall": 0.7583333333333333,
+      "eval_runtime": 2.091,
+      "eval_samples_per_second": 114.78,
+      "eval_steps_per_second": 3.826,
       "step": 40
     },
     {
       "epoch": 6.0,
-      "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,
+      "eval_accuracy": 0.7625,
+      "eval_f1_score": 0.7033555352126014,
+      "eval_loss": 0.5488410592079163,
+      "eval_precision": 0.7561481481481481,
+      "eval_recall": 0.7625,
+      "eval_runtime": 2.0904,
+      "eval_samples_per_second": 114.812,
+      "eval_steps_per_second": 3.827,
       "step": 48
     },
     {
       "epoch": 6.25,
       "learning_rate": 6e-06,
-      "loss": 0.5717,
+      "loss": 0.4893,
       "step": 50
     },
     {
       "epoch": 7.0,
-      "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,
+      "eval_accuracy": 0.7583333333333333,
+      "eval_f1_score": 0.7003333333333333,
+      "eval_loss": 0.5511656403541565,
+      "eval_precision": 0.7431547619047619,
+      "eval_recall": 0.7583333333333333,
+      "eval_runtime": 1.9003,
+      "eval_samples_per_second": 126.293,
+      "eval_steps_per_second": 4.21,
       "step": 56
     },
     {
       "epoch": 7.0,
       "step": 56,
       "total_flos": 1.73991922900992e+17,
-      "train_loss": 0.5857737362384796,
-      "train_runtime": 257.742,
-      "train_samples_per_second": 27.159,
-      "train_steps_per_second": 0.217
+      "train_loss": 0.5527358267988477,
+      "train_runtime": 138.7489,
+      "train_samples_per_second": 50.451,
+      "train_steps_per_second": 0.404
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7366666666666667,
-      "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,
+      "eval_accuracy": 0.78,
+      "eval_f1_score": 0.7315167493796526,
+      "eval_loss": 0.5118212699890137,
+      "eval_precision": 0.7896499764558155,
+      "eval_recall": 0.78,
+      "eval_runtime": 2.3258,
+      "eval_samples_per_second": 128.99,
+      "eval_steps_per_second": 4.3,
       "step": 56
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb7e8116dbe569fb96ff854b905cb0bc3ae5795528e7afede2c8328a68bddc7d
+oid sha256:8eba1d171a372a5066ba1cf45c0e4bcb11be97010d86e34d8bc234c36402be07
 size 4091