End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "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,
+    "epoch": 10.0,
+    "eval_accuracy": 0.8104042259990814,
+    "eval_f1": 0.8313368471490178,
+    "eval_loss": 0.5929840803146362,
+    "eval_precision": 0.7980680772769089,
+    "eval_recall": 0.8675,
+    "eval_runtime": 2.4572,
     "eval_samples": 54,
-    "eval_samples_per_second": 22.865,
-    "eval_steps_per_second": 2.541,
+    "eval_samples_per_second": 21.976,
+    "eval_steps_per_second": 2.442,
     "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": 0.2944237976074219,
-    "train_runtime": 336.0216,
+    "train_loss": 0.7810765075683593,
+    "train_runtime": 94.7164,
     "train_samples": 150,
-    "train_samples_per_second": 14.88,
-    "train_steps_per_second": 1.488
+    "train_samples_per_second": 15.837,
+    "train_steps_per_second": 1.584
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "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,
+    "epoch": 10.0,
+    "eval_accuracy": 0.8104042259990814,
+    "eval_f1": 0.8313368471490178,
+    "eval_loss": 0.5929840803146362,
+    "eval_precision": 0.7980680772769089,
+    "eval_recall": 0.8675,
+    "eval_runtime": 2.4572,
     "eval_samples": 54,
-    "eval_samples_per_second": 22.865,
-    "eval_steps_per_second": 2.541
+    "eval_samples_per_second": 21.976,
+    "eval_steps_per_second": 2.442
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:581afe274c45186d257308d437c3b39177f22b873e3cad10cde8017af1ae1d98
+oid sha256:841506f6c6c0a194bcb07157467ddcf167f6909db9a6af7260982b8d5abc6e08
 size 501420883

runs/Aug24_18-18-57_bernini/1724516347.7137752/events.out.tfevents.1724516347.bernini.11667.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58afa5d01e53942957c99fbb80b093bed7b1b20ce1a6dc0db13e7c827306a40f
+size 4671

runs/Aug24_18-18-57_bernini/events.out.tfevents.1724516347.bernini.11667.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10ee7afbe996f4a10db0be4b2ddbe3fb2f86ed738ee577f1b56295119c376792
+size 7229

runs/Aug24_18-18-57_bernini/events.out.tfevents.1724516450.bernini.11667.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bb9cbc7fbead01c5392d25c8908895a62506270486c77d4eaca2eb7edb3cd25
+size 512

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 33.33,
-    "train_loss": 0.2944237976074219,
-    "train_runtime": 336.0216,
+    "epoch": 10.0,
+    "train_loss": 0.7810765075683593,
+    "train_runtime": 94.7164,
     "train_samples": 150,
-    "train_samples_per_second": 14.88,
-    "train_steps_per_second": 1.488
+    "train_samples_per_second": 15.837,
+    "train_steps_per_second": 1.584
 }

trainer_state.json

@@ -1,151 +1,97 @@
 {
-  "best_metric": 0.8948016752894803,
-  "best_model_checkpoint": "/media/HDD/mpizzol/Tirocinio/layoutlmv3-base-finetuned-funsd/checkpoint-500",
-  "epoch": 33.333333333333336,
-  "global_step": 500,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "global_step": 150,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "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": 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": 1.67,
+      "eval_accuracy": 0.6559485530546624,
+      "eval_f1": 0.48850983179341384,
+      "eval_loss": 1.2209023237228394,
+      "eval_precision": 0.46420531292210715,
+      "eval_recall": 0.5155,
+      "eval_runtime": 2.3635,
+      "eval_samples_per_second": 22.848,
+      "eval_steps_per_second": 2.539,
+      "step": 25
     },
     {
-      "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": 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": 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": 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": 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": 3.33,
+      "eval_accuracy": 0.7619430408819476,
+      "eval_f1": 0.7535809662539452,
+      "eval_loss": 0.8171831965446472,
+      "eval_precision": 0.732420953279849,
+      "eval_recall": 0.776,
+      "eval_runtime": 2.3634,
+      "eval_samples_per_second": 22.849,
+      "eval_steps_per_second": 2.539,
+      "step": 50
     },
     {
-      "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": 5.0,
+      "eval_accuracy": 0.8125861276986679,
+      "eval_f1": 0.8145823273780782,
+      "eval_loss": 0.6125288605690002,
+      "eval_precision": 0.7875816993464052,
+      "eval_recall": 0.8435,
+      "eval_runtime": 2.3662,
+      "eval_samples_per_second": 22.821,
+      "eval_steps_per_second": 2.536,
+      "step": 75
     },
     {
-      "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": 6.67,
+      "eval_accuracy": 0.8107487367937529,
+      "eval_f1": 0.8347784200385356,
+      "eval_loss": 0.5983520150184631,
+      "eval_precision": 0.8052973977695167,
+      "eval_recall": 0.8665,
+      "eval_runtime": 2.4636,
+      "eval_samples_per_second": 21.92,
+      "eval_steps_per_second": 2.436,
+      "step": 100
     },
     {
-      "epoch": 33.33,
-      "learning_rate": 0.0,
-      "loss": 0.2944,
-      "step": 500
+      "epoch": 8.33,
+      "eval_accuracy": 0.8216582452916859,
+      "eval_f1": 0.8363723608445298,
+      "eval_loss": 0.5673876404762268,
+      "eval_precision": 0.8039667896678967,
+      "eval_recall": 0.8715,
+      "eval_runtime": 2.3684,
+      "eval_samples_per_second": 22.8,
+      "eval_steps_per_second": 2.533,
+      "step": 125
     },
     {
-      "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": 10.0,
+      "eval_accuracy": 0.8104042259990814,
+      "eval_f1": 0.8313368471490178,
+      "eval_loss": 0.5929840803146362,
+      "eval_precision": 0.7980680772769089,
+      "eval_recall": 0.8675,
+      "eval_runtime": 2.4505,
+      "eval_samples_per_second": 22.036,
+      "eval_steps_per_second": 2.448,
+      "step": 150
     },
     {
-      "epoch": 33.33,
-      "step": 500,
-      "total_flos": 1318026716160000.0,
-      "train_loss": 0.2944237976074219,
-      "train_runtime": 336.0216,
-      "train_samples_per_second": 14.88,
-      "train_steps_per_second": 1.488
+      "epoch": 10.0,
+      "step": 150,
+      "total_flos": 395408014848000.0,
+      "train_loss": 0.7810765075683593,
+      "train_runtime": 94.7164,
+      "train_samples_per_second": 15.837,
+      "train_steps_per_second": 1.584
     }
   ],
-  "max_steps": 500,
-  "num_train_epochs": 34,
-  "total_flos": 1318026716160000.0,
+  "max_steps": 150,
+  "num_train_epochs": 10,
+  "total_flos": 395408014848000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adc75eed38996e422651fb9e6e4b3f368310b1c9ad2de49c697d8cc53283e4b5
+oid sha256:67e671d048c6381947443975ece3fe1b7ffa48552b826dcbe8f9155c1dca716d
 size 2927