End of training

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 1.5988867282867432,
-  "best_model_checkpoint": "detr-r50-cd45rb-8ah-6l-gelu-corrected\\checkpoint-87514",
-  "epoch": 20.0,
-  "global_step": 92120,
+  "best_metric": 1.5861479043960571,
+  "best_model_checkpoint": "detr-r50-cd45rb-8ah-6l-gelu-corrected\\checkpoint-115150",
+  "epoch": 25.0,
+  "global_step": 115150,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -275,31 +275,101 @@
     },
     {
       "epoch": 20.0,
-      "learning_rate": 4.559270516717325e-09,
-      "loss": 2.0084,
+      "learning_rate": 2.0037342596613115e-06,
+      "loss": 2.0206,
       "step": 92120
     },
     {
       "epoch": 20.0,
-      "eval_loss": 1.599236011505127,
-      "eval_runtime": 202.2872,
-      "eval_samples_per_second": 8.804,
-      "eval_steps_per_second": 1.102,
+      "eval_loss": 1.6168476343154907,
+      "eval_runtime": 207.1893,
+      "eval_samples_per_second": 8.596,
+      "eval_steps_per_second": 1.076,
       "step": 92120
     },
     {
-      "epoch": 20.0,
-      "step": 92120,
-      "total_flos": 1.7606512956165125e+20,
-      "train_loss": 1.0303937374484369,
-      "train_runtime": 29465.5365,
-      "train_samples_per_second": 12.503,
-      "train_steps_per_second": 3.126
+      "epoch": 21.0,
+      "learning_rate": 1.6039947894051239e-06,
+      "loss": 2.026,
+      "step": 96726
+    },
+    {
+      "epoch": 21.0,
+      "eval_loss": 1.6022319793701172,
+      "eval_runtime": 206.9802,
+      "eval_samples_per_second": 8.605,
+      "eval_steps_per_second": 1.077,
+      "step": 96726
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 1.2041684759009988e-06,
+      "loss": 2.0109,
+      "step": 101332
+    },
+    {
+      "epoch": 22.0,
+      "eval_loss": 1.5995537042617798,
+      "eval_runtime": 206.8836,
+      "eval_samples_per_second": 8.609,
+      "eval_steps_per_second": 1.078,
+      "step": 101332
+    },
+    {
+      "epoch": 23.0,
+      "learning_rate": 8.042553191489362e-07,
+      "loss": 2.0133,
+      "step": 105938
+    },
+    {
+      "epoch": 23.0,
+      "eval_loss": 1.5983381271362305,
+      "eval_runtime": 206.4758,
+      "eval_samples_per_second": 8.626,
+      "eval_steps_per_second": 1.08,
+      "step": 105938
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 4.043421623968737e-07,
+      "loss": 2.0081,
+      "step": 110544
+    },
+    {
+      "epoch": 24.0,
+      "eval_loss": 1.5887646675109863,
+      "eval_runtime": 203.9191,
+      "eval_samples_per_second": 8.734,
+      "eval_steps_per_second": 1.094,
+      "step": 110544
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 4.515848892748589e-09,
+      "loss": 1.9975,
+      "step": 115150
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 1.5861479043960571,
+      "eval_runtime": 204.8681,
+      "eval_samples_per_second": 8.693,
+      "eval_steps_per_second": 1.089,
+      "step": 115150
+    },
+    {
+      "epoch": 25.0,
+      "step": 115150,
+      "total_flos": 2.2008141195206407e+20,
+      "train_loss": 0.4830571452317629,
+      "train_runtime": 17891.958,
+      "train_samples_per_second": 25.739,
+      "train_steps_per_second": 6.436
     }
   ],
-  "max_steps": 92120,
-  "num_train_epochs": 20,
-  "total_flos": 1.7606512956165125e+20,
+  "max_steps": 115150,
+  "num_train_epochs": 25,
+  "total_flos": 2.2008141195206407e+20,
   "trial_name": null,
   "trial_params": null
 }