End of training

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 1.8535605669021606,
-  "best_model_checkpoint": "detr-r101-cd45rb-8ah-4l\\checkpoint-46060",
-  "epoch": 10.0,
-  "global_step": 46060,
+  "best_metric": 1.7442328929901123,
+  "best_model_checkpoint": "detr-r101-cd45rb-8ah-4l\\checkpoint-115150",
+  "epoch": 25.0,
+  "global_step": 115150,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -148,18 +148,228 @@
       "step": 46060
     },
     {
-      "epoch": 10.0,
-      "step": 46060,
-      "total_flos": 1.1594295671662082e+20,
-      "train_loss": 2.5832694235440186,
-      "train_runtime": 33090.2205,
-      "train_samples_per_second": 5.567,
-      "train_steps_per_second": 1.392
+      "epoch": 11.0,
+      "learning_rate": 5.60243161094225e-06,
+      "loss": 2.4311,
+      "step": 50666
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 1.8880857229232788,
+      "eval_runtime": 215.2412,
+      "eval_samples_per_second": 8.274,
+      "eval_steps_per_second": 1.036,
+      "step": 50666
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 5.202518454190188e-06,
+      "loss": 2.4175,
+      "step": 55272
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 1.844258189201355,
+      "eval_runtime": 215.4657,
+      "eval_samples_per_second": 8.266,
+      "eval_steps_per_second": 1.035,
+      "step": 55272
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 4.802605297438124e-06,
+      "loss": 2.3763,
+      "step": 59878
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 1.8380271196365356,
+      "eval_runtime": 216.3098,
+      "eval_samples_per_second": 8.234,
+      "eval_steps_per_second": 1.031,
+      "step": 59878
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 4.402692140686062e-06,
+      "loss": 2.371,
+      "step": 64484
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 1.831394076347351,
+      "eval_runtime": 217.0583,
+      "eval_samples_per_second": 8.205,
+      "eval_steps_per_second": 1.027,
+      "step": 64484
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 4.002865827181937e-06,
+      "loss": 2.3427,
+      "step": 69090
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 1.8254896402359009,
+      "eval_runtime": 216.9087,
+      "eval_samples_per_second": 8.211,
+      "eval_steps_per_second": 1.028,
+      "step": 69090
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 3.603126356925749e-06,
+      "loss": 2.3415,
+      "step": 73696
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 1.8258094787597656,
+      "eval_runtime": 217.3195,
+      "eval_samples_per_second": 8.195,
+      "eval_steps_per_second": 1.026,
+      "step": 73696
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.2032132001736863e-06,
+      "loss": 2.3189,
+      "step": 78302
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 1.7938063144683838,
+      "eval_runtime": 217.2158,
+      "eval_samples_per_second": 8.199,
+      "eval_steps_per_second": 1.027,
+      "step": 78302
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 2.803300043421624e-06,
+      "loss": 2.3049,
+      "step": 82908
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 1.7958720922470093,
+      "eval_runtime": 217.0427,
+      "eval_samples_per_second": 8.206,
+      "eval_steps_per_second": 1.027,
+      "step": 82908
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 2.403473729917499e-06,
+      "loss": 2.2884,
+      "step": 87514
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 1.7826542854309082,
+      "eval_runtime": 216.689,
+      "eval_samples_per_second": 8.219,
+      "eval_steps_per_second": 1.029,
+      "step": 87514
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 2.0037342596613115e-06,
+      "loss": 2.2769,
+      "step": 92120
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 1.7849496603012085,
+      "eval_runtime": 216.0142,
+      "eval_samples_per_second": 8.245,
+      "eval_steps_per_second": 1.032,
+      "step": 92120
+    },
+    {
+      "epoch": 21.0,
+      "learning_rate": 1.6040816326530614e-06,
+      "loss": 2.2728,
+      "step": 96726
+    },
+    {
+      "epoch": 21.0,
+      "eval_loss": 1.7580715417861938,
+      "eval_runtime": 218.4831,
+      "eval_samples_per_second": 8.152,
+      "eval_steps_per_second": 1.021,
+      "step": 96726
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 1.2041684759009988e-06,
+      "loss": 2.2547,
+      "step": 101332
+    },
+    {
+      "epoch": 22.0,
+      "eval_loss": 1.7552635669708252,
+      "eval_runtime": 216.3596,
+      "eval_samples_per_second": 8.232,
+      "eval_steps_per_second": 1.031,
+      "step": 101332
+    },
+    {
+      "epoch": 23.0,
+      "learning_rate": 8.041684759009987e-07,
+      "loss": 2.249,
+      "step": 105938
+    },
+    {
+      "epoch": 23.0,
+      "eval_loss": 1.7520817518234253,
+      "eval_runtime": 215.9835,
+      "eval_samples_per_second": 8.246,
+      "eval_steps_per_second": 1.032,
+      "step": 105938
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 4.043421623968737e-07,
+      "loss": 2.2476,
+      "step": 110544
+    },
+    {
+      "epoch": 24.0,
+      "eval_loss": 1.7504931688308716,
+      "eval_runtime": 215.9138,
+      "eval_samples_per_second": 8.249,
+      "eval_steps_per_second": 1.033,
+      "step": 110544
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 4.429005644811116e-09,
+      "loss": 2.2371,
+      "step": 115150
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 1.7442328929901123,
+      "eval_runtime": 218.3812,
+      "eval_samples_per_second": 8.155,
+      "eval_steps_per_second": 1.021,
+      "step": 115150
+    },
+    {
+      "epoch": 25.0,
+      "step": 115150,
+      "total_flos": 2.8985739179155207e+20,
+      "train_loss": 1.3892145011262484,
+      "train_runtime": 49656.2093,
+      "train_samples_per_second": 9.274,
+      "train_steps_per_second": 2.319
     }
   ],
-  "max_steps": 46060,
-  "num_train_epochs": 10,
-  "total_flos": 1.1594295671662082e+20,
+  "max_steps": 115150,
+  "num_train_epochs": 25,
+  "total_flos": 2.8985739179155207e+20,
   "trial_name": null,
   "trial_params": null
 }