Training in progress, step 400, checkpoint

checkpoint-400/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-400/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f32caa3ade0bc4dd79232fd1456fae9603bb7c02292df8675852e005eff55e9
+oid sha256:1e8dea633ddefd5c6a77825a5ad48949cacf27d7dfd7be23bdb43604b3445369
 size 1134332228

checkpoint-400/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e18282059758f7438f8035104343961dae0354a3bf096166da6d8f2c8402bc4a
-size 2265924830
+oid sha256:a51c90031d2351c0172e0d11461c02d6a3927f46e41e12851994a1e424524116
+size 2268900866

checkpoint-400/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:662e921e03fcf92563a3b2458b6412a5d7887e0762378509ed7fc9f2b0e4bae7
+oid sha256:9f5daab7d556fc02d85dffe50c326709842df42c35ef620127cae6836a494e7c
 size 14244

checkpoint-400/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d2d6ad579de8c53cc7768e169f4897cd81cf9311331caed421754d374415e90
+oid sha256:02ca0267ce9457847149d1dc514b449ab61a93b118b8048db3838e7a156f3daa
 size 1064

checkpoint-400/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:683eb7c7eca2f8d4661f33cb19ad5dcb7c826bbc4256f314d42c7d9d3ac93240
-size 17083001
+oid sha256:3a56def25aa40facc030ea8b0b87f3688e4b3c39eb8b45d5702b3a1300fe2a20
+size 17082734

checkpoint-400/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7865890784233485,
-  "best_model_checkpoint": "DocLayNet/lilt-xlm-roberta-base-finetuned-DocLayNet-base_paragraphs_ml512-v1/checkpoint-300",
+  "best_metric": 0.8202860650583409,
+  "best_model_checkpoint": "DocLayNet/lilt-xlm-roberta-base-finetuned-DocLayNet-base_paragraphs_ml512-v1/checkpoint-200",
   "epoch": 0.21321961620469082,
   "eval_steps": 100,
   "global_step": 400,
@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 0.053304904051172705,
-      "eval_accuracy": 0.6912373860880154,
-      "eval_f1": 0.6912373860880154,
-      "eval_loss": 0.8656882047653198,
-      "eval_precision": 0.6912373860880154,
-      "eval_recall": 0.6912373860880154,
-      "eval_runtime": 41.8245,
-      "eval_samples_per_second": 38.422,
-      "eval_steps_per_second": 2.415,
+      "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": 0.7121482827113298,
-      "eval_f1": 0.7121482827113298,
-      "eval_loss": 0.856532096862793,
-      "eval_precision": 0.7121482827113298,
-      "eval_recall": 0.7121482827113298,
-      "eval_runtime": 42.6291,
-      "eval_samples_per_second": 37.697,
-      "eval_steps_per_second": 2.369,
+      "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": 0.7865890784233484,
-      "eval_f1": 0.7865890784233485,
-      "eval_loss": 0.6567804217338562,
-      "eval_precision": 0.7865890784233484,
-      "eval_recall": 0.7865890784233484,
-      "eval_runtime": 41.1268,
-      "eval_samples_per_second": 39.074,
-      "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
     },
     {
       "epoch": 0.21321961620469082,
-      "eval_accuracy": 0.7400361324323516,
-      "eval_f1": 0.7400361324323516,
-      "eval_loss": 0.575270414352417,
-      "eval_precision": 0.7400361324323516,
-      "eval_recall": 0.7400361324323516,
-      "eval_runtime": 43.0681,
-      "eval_samples_per_second": 37.313,
-      "eval_steps_per_second": 2.345,
+      "eval_accuracy": 0.7694710890634513,
+      "eval_f1": 0.7694710890634513,
+      "eval_loss": 0.5369378328323364,
+      "eval_precision": 0.7694710890634513,
+      "eval_recall": 0.7694710890634513,
+      "eval_runtime": 69.0856,
+      "eval_samples_per_second": 23.261,
+      "eval_steps_per_second": 0.738,
       "step": 400
     }
   ],
@@ -75,7 +75,7 @@
     }
   },
   "total_flos": 890151164313600.0,
-  "train_batch_size": 8,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-400/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d884d25b9a0a1975c8ad8cdce24ef6b7c329f16153a80771adc673a879e2c8c3
+oid sha256:2544ef83614d223e3dbe1cbe496bce99a66932dd46abc19142f548bbc4c106c4
 size 5304