Training in progress, step 300, checkpoint

checkpoint-300/config.json

@@ -49,7 +49,7 @@
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.42.4",
+  "transformers_version": "4.42.3",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 250002

checkpoint-300/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8de3c8c8f3a40d30b88eda3ccdc2516425d7d0dc954f63b9f66d81a7558ef54f
+oid sha256:de73324ea872c24355ab16485873b5028f9a4d7e45c5ece93ceae2f74575c683
 size 1134332228

checkpoint-300/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cc47348c702a0022af13c041848d24df2c6443e1bb3c48fd00e490f3319bb92
-size 2265924830
+oid sha256:2130478a54e077cfc19dcf4cd2cdb8cf72095ad0d98211da1a2a7c3a817ded2e
+size 2268900866

checkpoint-300/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc421c8fddea08151d7ad39805822c2e21285c984f139055f5044a28cf0c6db0
+oid sha256:0d371bd4c74e7ce6f5c1d9f813a146b7ff337a970d2b013cb2801da88029b05e
 size 14244

checkpoint-300/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52994157dc052c931eef58e1d32a903316896aaa193ff6d3c5495b1a13852974
+oid sha256:8f8f38a4f7c169432a2871ed6f000991f28807d2b3ca05509f84300c0489f407
 size 1064

checkpoint-300/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7496182139399324,
+  "best_metric": 0.8202860650583409,
   "best_model_checkpoint": "DocLayNet/lilt-xlm-roberta-base-finetuned-DocLayNet-base_paragraphs_ml512-v1/checkpoint-200",
   "epoch": 0.15991471215351813,
   "eval_steps": 100,
@@ -10,38 +10,38 @@
   "log_history": [
     {
       "epoch": 0.053304904051172705,
-      "eval_accuracy": 0.7157016379370577,
-      "eval_f1": 0.7157016379370577,
-      "eval_loss": 0.8210025429725647,
-      "eval_precision": 0.7157016379370577,
-      "eval_recall": 0.7157016379370577,
-      "eval_runtime": 40.0037,
-      "eval_samples_per_second": 40.171,
-      "eval_steps_per_second": 2.525,
+      "eval_accuracy": 0.6321578647928374,
+      "eval_f1": 0.6321578647928374,
+      "eval_loss": 0.9594171643257141,
+      "eval_precision": 0.6321578647928374,
+      "eval_recall": 0.6321578647928374,
+      "eval_runtime": 69.4219,
+      "eval_samples_per_second": 23.148,
+      "eval_steps_per_second": 0.735,
       "step": 100
     },
     {
       "epoch": 0.10660980810234541,
-      "eval_accuracy": 0.7496182139399323,
-      "eval_f1": 0.7496182139399324,
-      "eval_loss": 0.673028290271759,
-      "eval_precision": 0.7496182139399323,
-      "eval_recall": 0.7496182139399323,
-      "eval_runtime": 42.4041,
-      "eval_samples_per_second": 37.897,
-      "eval_steps_per_second": 2.382,
+      "eval_accuracy": 0.8202860650583409,
+      "eval_f1": 0.8202860650583409,
+      "eval_loss": 0.6088714599609375,
+      "eval_precision": 0.8202860650583409,
+      "eval_recall": 0.8202860650583409,
+      "eval_runtime": 69.0048,
+      "eval_samples_per_second": 23.288,
+      "eval_steps_per_second": 0.739,
       "step": 200
     },
     {
       "epoch": 0.15991471215351813,
-      "eval_accuracy": 0.7356044197350954,
-      "eval_f1": 0.7356044197350955,
-      "eval_loss": 0.6603901982307434,
-      "eval_precision": 0.7356044197350954,
-      "eval_recall": 0.7356044197350954,
-      "eval_runtime": 41.1176,
-      "eval_samples_per_second": 39.083,
-      "eval_steps_per_second": 2.456,
+      "eval_accuracy": 0.7378602014233383,
+      "eval_f1": 0.7378602014233383,
+      "eval_loss": 0.7009087800979614,
+      "eval_precision": 0.7378602014233383,
+      "eval_recall": 0.7378602014233383,
+      "eval_runtime": 68.7975,
+      "eval_samples_per_second": 23.358,
+      "eval_steps_per_second": 0.741,
       "step": 300
     }
   ],
@@ -62,8 +62,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 890151164313600.0,
-  "train_batch_size": 8,
+  "total_flos": 667613373235200.0,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-300/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:623a05d696ac4fa501371203f02f03749bacbe9ec22f5e54f15d14e9636da808
+oid sha256:2544ef83614d223e3dbe1cbe496bce99a66932dd46abc19142f548bbc4c106c4
 size 5304