End of training

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.9384550084889643,
+    "eval_f1": 0.9269016697588126,
+    "eval_loss": 0.30563604831695557,
+    "eval_precision": 0.9190581309786607,
+    "eval_recall": 0.9348802395209581,
+    "eval_runtime": 8.103,
+    "eval_samples": 100,
+    "eval_samples_per_second": 12.341,
+    "eval_steps_per_second": 0.864,
+    "train_loss": 0.7290411071777344,
+    "train_runtime": 1777.9741,
+    "train_samples": 800,
+    "train_samples_per_second": 8.999,
+    "train_steps_per_second": 0.562
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.9384550084889643,
+    "eval_f1": 0.9269016697588126,
+    "eval_loss": 0.30563604831695557,
+    "eval_precision": 0.9190581309786607,
+    "eval_recall": 0.9348802395209581,
+    "eval_runtime": 8.103,
+    "eval_samples": 100,
+    "eval_samples_per_second": 12.341,
+    "eval_steps_per_second": 0.864
+}

runs/May02_17-47-48_brutasse/events.out.tfevents.1651516292.brutasse.19888.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9de232cdbaea82115cd76733b0d0fcd778eea43446f40a50892537f4d5504d02
+size 512

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 0.7290411071777344,
+    "train_runtime": 1777.9741,
+    "train_samples": 800,
+    "train_samples_per_second": 8.999,
+    "train_steps_per_second": 0.562
+}

trainer_state.json

@@ -0,0 +1,157 @@
+{
+  "best_metric": 0.9269016697588126,
+  "best_model_checkpoint": "layoutlmv3-test-cord/checkpoint-1000",
+  "epoch": 20.0,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6366723259762309,
+      "eval_f1": 0.5626932325661285,
+      "eval_loss": 1.6054191589355469,
+      "eval_precision": 0.52,
+      "eval_recall": 0.6130239520958084,
+      "eval_runtime": 8.1419,
+      "eval_samples_per_second": 12.282,
+      "eval_steps_per_second": 0.86,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8314940577249575,
+      "eval_f1": 0.8096632503660322,
+      "eval_loss": 0.9171701669692993,
+      "eval_precision": 0.7922636103151862,
+      "eval_recall": 0.8278443113772455,
+      "eval_runtime": 8.1716,
+      "eval_samples_per_second": 12.238,
+      "eval_steps_per_second": 0.857,
+      "step": 200
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.866723259762309,
+      "eval_f1": 0.8496683861459101,
+      "eval_loss": 0.6381942629814148,
+      "eval_precision": 0.8367198838896952,
+      "eval_recall": 0.8630239520958084,
+      "eval_runtime": 8.0786,
+      "eval_samples_per_second": 12.378,
+      "eval_steps_per_second": 0.866,
+      "step": 300
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8960101867572157,
+      "eval_f1": 0.8775811209439529,
+      "eval_loss": 0.4974280595779419,
+      "eval_precision": 0.8648255813953488,
+      "eval_recall": 0.8907185628742516,
+      "eval_runtime": 8.0768,
+      "eval_samples_per_second": 12.381,
+      "eval_steps_per_second": 0.867,
+      "step": 400
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 5e-06,
+      "loss": 1.1589,
+      "step": 500
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.916383701188455,
+      "eval_f1": 0.8914243651085757,
+      "eval_loss": 0.41238266229629517,
+      "eval_precision": 0.8769007965242578,
+      "eval_recall": 0.906437125748503,
+      "eval_runtime": 8.1158,
+      "eval_samples_per_second": 12.322,
+      "eval_steps_per_second": 0.863,
+      "step": 500
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.9235993208828522,
+      "eval_f1": 0.9064002959674436,
+      "eval_loss": 0.37666961550712585,
+      "eval_precision": 0.8961228968544257,
+      "eval_recall": 0.9169161676646707,
+      "eval_runtime": 8.1123,
+      "eval_samples_per_second": 12.327,
+      "eval_steps_per_second": 0.863,
+      "step": 600
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.933786078098472,
+      "eval_f1": 0.9210818821785848,
+      "eval_loss": 0.3387647271156311,
+      "eval_precision": 0.9119589141599413,
+      "eval_recall": 0.9303892215568862,
+      "eval_runtime": 8.0833,
+      "eval_samples_per_second": 12.371,
+      "eval_steps_per_second": 0.866,
+      "step": 700
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.9393039049235993,
+      "eval_f1": 0.927643784786642,
+      "eval_loss": 0.3137519955635071,
+      "eval_precision": 0.919793966151582,
+      "eval_recall": 0.9356287425149701,
+      "eval_runtime": 7.8384,
+      "eval_samples_per_second": 12.758,
+      "eval_steps_per_second": 0.893,
+      "step": 800
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.9376061120543294,
+      "eval_f1": 0.9254174397031539,
+      "eval_loss": 0.30727389454841614,
+      "eval_precision": 0.9175864606328182,
+      "eval_recall": 0.9333832335329342,
+      "eval_runtime": 7.8694,
+      "eval_samples_per_second": 12.708,
+      "eval_steps_per_second": 0.89,
+      "step": 900
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0,
+      "loss": 0.2992,
+      "step": 1000
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.9384550084889643,
+      "eval_f1": 0.9269016697588126,
+      "eval_loss": 0.30563604831695557,
+      "eval_precision": 0.9190581309786607,
+      "eval_recall": 0.9348802395209581,
+      "eval_runtime": 7.9092,
+      "eval_samples_per_second": 12.644,
+      "eval_steps_per_second": 0.885,
+      "step": 1000
+    },
+    {
+      "epoch": 20.0,
+      "step": 1000,
+      "total_flos": 4248755355648000.0,
+      "train_loss": 0.7290411071777344,
+      "train_runtime": 1777.9741,
+      "train_samples_per_second": 8.999,
+      "train_steps_per_second": 0.562
+    }
+  ],
+  "max_steps": 1000,
+  "num_train_epochs": 20,
+  "total_flos": 4248755355648000.0,
+  "trial_name": null,
+  "trial_params": null
+}