End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 13.16,
-    "eval_accuracy": 0.843706936150666,
-    "eval_f1": 0.8903638151425762,
-    "eval_loss": 0.6187416911125183,
-    "eval_precision": 0.875725338491296,
-    "eval_recall": 0.9055,
-    "eval_runtime": 2.2631,
+    "epoch": 33.33,
+    "eval_accuracy": 0.8486449242076252,
+    "eval_f1": 0.8948016752894803,
+    "eval_loss": 0.6415959000587463,
+    "eval_precision": 0.8819815444390481,
+    "eval_recall": 0.908,
+    "eval_runtime": 2.3617,
     "eval_samples": 54,
-    "eval_samples_per_second": 23.861,
-    "eval_steps_per_second": 1.767,
+    "eval_samples_per_second": 22.865,
+    "eval_steps_per_second": 2.541,
     "predict_accuracy": 0.8218373936014088,
     "predict_f1": 0.9038133181559477,
     "predict_loss": 0.6888472437858582,
@@ -17,9 +17,9 @@
     "predict_runtime": 2.0596,
     "predict_samples_per_second": 8.74,
     "predict_steps_per_second": 0.971,
-    "train_loss": 0.43045907592773436,
-    "train_runtime": 242.2202,
+    "train_loss": 0.2944237976074219,
+    "train_runtime": 336.0216,
     "train_samples": 150,
-    "train_samples_per_second": 8.257,
-    "train_steps_per_second": 2.064
+    "train_samples_per_second": 14.88,
+    "train_steps_per_second": 1.488
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 13.16,
-    "eval_accuracy": 0.843706936150666,
-    "eval_f1": 0.8903638151425762,
-    "eval_loss": 0.6187416911125183,
-    "eval_precision": 0.875725338491296,
-    "eval_recall": 0.9055,
-    "eval_runtime": 2.2631,
+    "epoch": 33.33,
+    "eval_accuracy": 0.8486449242076252,
+    "eval_f1": 0.8948016752894803,
+    "eval_loss": 0.6415959000587463,
+    "eval_precision": 0.8819815444390481,
+    "eval_recall": 0.908,
+    "eval_runtime": 2.3617,
     "eval_samples": 54,
-    "eval_samples_per_second": 23.861,
-    "eval_steps_per_second": 1.767
+    "eval_samples_per_second": 22.865,
+    "eval_steps_per_second": 2.541
 }

runs/Aug24_18-11-27_bernini/events.out.tfevents.1724515898.bernini.30822.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f315bfd7bfb7068843ca3065ef7e4b272ac8e0b665c0f86a64983c3ed2927b99
-size 8947
+oid sha256:17b2a58640434ba148ade9740cc6b9418f9927cca8b51fc5ff9e64b4457e5ab7
+size 9301

runs/Aug24_18-11-27_bernini/events.out.tfevents.1724516241.bernini.30822.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af1c8a8e01bba8de37f01cd35f53370abe6ac9ca8474c172909a45ecc0b01c74
+size 512

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 13.16,
-    "train_loss": 0.43045907592773436,
-    "train_runtime": 242.2202,
+    "epoch": 33.33,
+    "train_loss": 0.2944237976074219,
+    "train_runtime": 336.0216,
     "train_samples": 150,
-    "train_samples_per_second": 8.257,
-    "train_steps_per_second": 2.064
+    "train_samples_per_second": 14.88,
+    "train_steps_per_second": 1.488
 }

trainer_state.json

@@ -1,151 +1,151 @@
 {
-  "best_metric": 0.8903638151425762,
+  "best_metric": 0.8948016752894803,
   "best_model_checkpoint": "/media/HDD/mpizzol/Tirocinio/layoutlmv3-base-finetuned-funsd/checkpoint-500",
-  "epoch": 13.157894736842104,
+  "epoch": 33.333333333333336,
   "global_step": 500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.32,
-      "eval_accuracy": 0.7606798346348186,
-      "eval_f1": 0.7277096851718338,
-      "eval_loss": 0.9062727689743042,
-      "eval_precision": 0.7006015733456733,
-      "eval_recall": 0.757,
-      "eval_runtime": 2.4285,
-      "eval_samples_per_second": 22.236,
-      "eval_steps_per_second": 1.647,
+      "epoch": 3.33,
+      "eval_accuracy": 0.7633210840606339,
+      "eval_f1": 0.78394616678683,
+      "eval_loss": 0.7809513211250305,
+      "eval_precision": 0.7547431744562703,
+      "eval_recall": 0.8155,
+      "eval_runtime": 2.4166,
+      "eval_samples_per_second": 22.346,
+      "eval_steps_per_second": 2.483,
       "step": 50
     },
     {
-      "epoch": 2.63,
-      "eval_accuracy": 0.7967386311437759,
-      "eval_f1": 0.8242074927953891,
-      "eval_loss": 0.6386759281158447,
-      "eval_precision": 0.7929759704251387,
-      "eval_recall": 0.858,
-      "eval_runtime": 2.2823,
-      "eval_samples_per_second": 23.66,
-      "eval_steps_per_second": 1.753,
+      "epoch": 6.67,
+      "eval_accuracy": 0.8060404225999082,
+      "eval_f1": 0.8391708363116512,
+      "eval_loss": 0.5576171278953552,
+      "eval_precision": 0.8015475648611743,
+      "eval_recall": 0.8805,
+      "eval_runtime": 2.4186,
+      "eval_samples_per_second": 22.327,
+      "eval_steps_per_second": 2.481,
       "step": 100
     },
     {
-      "epoch": 3.95,
-      "eval_accuracy": 0.825447864033073,
-      "eval_f1": 0.8485576923076923,
-      "eval_loss": 0.5691221356391907,
-      "eval_precision": 0.8171296296296297,
-      "eval_recall": 0.8825,
-      "eval_runtime": 2.2496,
-      "eval_samples_per_second": 24.004,
-      "eval_steps_per_second": 1.778,
+      "epoch": 10.0,
+      "eval_accuracy": 0.822347266881029,
+      "eval_f1": 0.8655769699926811,
+      "eval_loss": 0.5810444355010986,
+      "eval_precision": 0.845164363982849,
+      "eval_recall": 0.887,
+      "eval_runtime": 2.3446,
+      "eval_samples_per_second": 23.032,
+      "eval_steps_per_second": 2.559,
       "step": 150
     },
     {
-      "epoch": 5.26,
-      "eval_accuracy": 0.822347266881029,
-      "eval_f1": 0.8555474629764507,
-      "eval_loss": 0.5723332166671753,
-      "eval_precision": 0.831524303916942,
-      "eval_recall": 0.881,
-      "eval_runtime": 2.2738,
-      "eval_samples_per_second": 23.749,
-      "eval_steps_per_second": 1.759,
+      "epoch": 13.33,
+      "eval_accuracy": 0.8393431327514929,
+      "eval_f1": 0.8725060827250608,
+      "eval_loss": 0.5633729100227356,
+      "eval_precision": 0.8497630331753554,
+      "eval_recall": 0.8965,
+      "eval_runtime": 2.3645,
+      "eval_samples_per_second": 22.837,
+      "eval_steps_per_second": 2.537,
       "step": 200
     },
     {
-      "epoch": 6.58,
-      "eval_accuracy": 0.8292374827744603,
-      "eval_f1": 0.8729388942774006,
-      "eval_loss": 0.5896942019462585,
-      "eval_precision": 0.847457627118644,
-      "eval_recall": 0.9,
-      "eval_runtime": 2.2927,
-      "eval_samples_per_second": 23.553,
-      "eval_steps_per_second": 1.745,
+      "epoch": 16.67,
+      "eval_accuracy": 0.8528938906752411,
+      "eval_f1": 0.8940364711680631,
+      "eval_loss": 0.5418661236763,
+      "eval_precision": 0.8814382896015549,
+      "eval_recall": 0.907,
+      "eval_runtime": 2.3455,
+      "eval_samples_per_second": 23.023,
+      "eval_steps_per_second": 2.558,
       "step": 250
     },
     {
-      "epoch": 7.89,
-      "eval_accuracy": 0.8283187873220027,
-      "eval_f1": 0.874515503875969,
-      "eval_loss": 0.6122345924377441,
-      "eval_precision": 0.8482142857142857,
-      "eval_recall": 0.9025,
-      "eval_runtime": 2.2852,
-      "eval_samples_per_second": 23.63,
-      "eval_steps_per_second": 1.75,
+      "epoch": 20.0,
+      "eval_accuracy": 0.8464630225080386,
+      "eval_f1": 0.8880670611439841,
+      "eval_loss": 0.5817353129386902,
+      "eval_precision": 0.8759727626459144,
+      "eval_recall": 0.9005,
+      "eval_runtime": 2.3522,
+      "eval_samples_per_second": 22.957,
+      "eval_steps_per_second": 2.551,
       "step": 300
     },
     {
-      "epoch": 9.21,
-      "eval_accuracy": 0.8392282958199357,
-      "eval_f1": 0.8740884783665532,
-      "eval_loss": 0.6045289635658264,
-      "eval_precision": 0.8505203405865658,
-      "eval_recall": 0.899,
-      "eval_runtime": 2.4063,
-      "eval_samples_per_second": 22.441,
-      "eval_steps_per_second": 1.662,
+      "epoch": 23.33,
+      "eval_accuracy": 0.8429030776297657,
+      "eval_f1": 0.8911230995586071,
+      "eval_loss": 0.6015072464942932,
+      "eval_precision": 0.8743984600577478,
+      "eval_recall": 0.9085,
+      "eval_runtime": 2.3531,
+      "eval_samples_per_second": 22.948,
+      "eval_steps_per_second": 2.55,
       "step": 350
     },
     {
-      "epoch": 10.53,
-      "eval_accuracy": 0.8446256316031235,
-      "eval_f1": 0.8851733464470125,
-      "eval_loss": 0.5662067532539368,
-      "eval_precision": 0.8708272859216255,
-      "eval_recall": 0.9,
-      "eval_runtime": 2.2543,
-      "eval_samples_per_second": 23.954,
-      "eval_steps_per_second": 1.774,
+      "epoch": 26.67,
+      "eval_accuracy": 0.8535829122645843,
+      "eval_f1": 0.8996811380917341,
+      "eval_loss": 0.5981678366661072,
+      "eval_precision": 0.8830043331728454,
+      "eval_recall": 0.917,
+      "eval_runtime": 2.4921,
+      "eval_samples_per_second": 21.669,
+      "eval_steps_per_second": 2.408,
       "step": 400
     },
     {
-      "epoch": 11.84,
-      "eval_accuracy": 0.843706936150666,
-      "eval_f1": 0.888943488943489,
-      "eval_loss": 0.5973214507102966,
-      "eval_precision": 0.8739130434782608,
-      "eval_recall": 0.9045,
-      "eval_runtime": 2.2748,
-      "eval_samples_per_second": 23.738,
-      "eval_steps_per_second": 1.758,
+      "epoch": 30.0,
+      "eval_accuracy": 0.8493339457969683,
+      "eval_f1": 0.8949185989146522,
+      "eval_loss": 0.6316392421722412,
+      "eval_precision": 0.8831548198636806,
+      "eval_recall": 0.907,
+      "eval_runtime": 2.359,
+      "eval_samples_per_second": 22.891,
+      "eval_steps_per_second": 2.543,
       "step": 450
     },
     {
-      "epoch": 13.16,
+      "epoch": 33.33,
       "learning_rate": 0.0,
-      "loss": 0.4305,
+      "loss": 0.2944,
       "step": 500
     },
     {
-      "epoch": 13.16,
-      "eval_accuracy": 0.843706936150666,
-      "eval_f1": 0.8903638151425762,
-      "eval_loss": 0.6187416911125183,
-      "eval_precision": 0.875725338491296,
-      "eval_recall": 0.9055,
-      "eval_runtime": 2.216,
-      "eval_samples_per_second": 24.369,
-      "eval_steps_per_second": 1.805,
+      "epoch": 33.33,
+      "eval_accuracy": 0.8486449242076252,
+      "eval_f1": 0.8948016752894803,
+      "eval_loss": 0.6415959000587463,
+      "eval_precision": 0.8819815444390481,
+      "eval_recall": 0.908,
+      "eval_runtime": 2.3662,
+      "eval_samples_per_second": 22.821,
+      "eval_steps_per_second": 2.536,
       "step": 500
     },
     {
-      "epoch": 13.16,
+      "epoch": 33.33,
       "step": 500,
-      "total_flos": 520356947539968.0,
-      "train_loss": 0.43045907592773436,
-      "train_runtime": 242.2202,
-      "train_samples_per_second": 8.257,
-      "train_steps_per_second": 2.064
+      "total_flos": 1318026716160000.0,
+      "train_loss": 0.2944237976074219,
+      "train_runtime": 336.0216,
+      "train_samples_per_second": 14.88,
+      "train_steps_per_second": 1.488
     }
   ],
   "max_steps": 500,
-  "num_train_epochs": 14,
-  "total_flos": 520356947539968.0,
+  "num_train_epochs": 34,
+  "total_flos": 1318026716160000.0,
   "trial_name": null,
   "trial_params": null
 }