Training in progress, step 5100

checkpoint-5100/config.json

@@ -0,0 +1,161 @@
+{
+  "_name_or_path": "microsoft/layoutxlm-base",
+  "architectures": [
+    "LayoutLMv2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "convert_sync_batchnorm": true,
+  "coordinate_size": 128,
+  "detectron2_config_args": {
+    "MODEL.ANCHOR_GENERATOR.SIZES": [
+      [
+        32
+      ],
+      [
+        64
+      ],
+      [
+        128
+      ],
+      [
+        256
+      ],
+      [
+        512
+      ]
+    ],
+    "MODEL.BACKBONE.NAME": "build_resnet_fpn_backbone",
+    "MODEL.FPN.IN_FEATURES": [
+      "res2",
+      "res3",
+      "res4",
+      "res5"
+    ],
+    "MODEL.MASK_ON": true,
+    "MODEL.PIXEL_STD": [
+      57.375,
+      57.12,
+      58.395
+    ],
+    "MODEL.POST_NMS_TOPK_TEST": 1000,
+    "MODEL.RESNETS.ASPECT_RATIOS": [
+      [
+        0.5,
+        1.0,
+        2.0
+      ]
+    ],
+    "MODEL.RESNETS.DEPTH": 101,
+    "MODEL.RESNETS.NUM_GROUPS": 32,
+    "MODEL.RESNETS.OUT_FEATURES": [
+      "res2",
+      "res3",
+      "res4",
+      "res5"
+    ],
+    "MODEL.RESNETS.SIZES": [
+      [
+        32
+      ],
+      [
+        64
+      ],
+      [
+        128
+      ],
+      [
+        256
+      ],
+      [
+        512
+      ]
+    ],
+    "MODEL.RESNETS.STRIDE_IN_1X1": false,
+    "MODEL.RESNETS.WIDTH_PER_GROUP": 8,
+    "MODEL.ROI_BOX_HEAD.NAME": "FastRCNNConvFCHead",
+    "MODEL.ROI_BOX_HEAD.NUM_FC": 2,
+    "MODEL.ROI_BOX_HEAD.POOLER_RESOLUTION": 14,
+    "MODEL.ROI_HEADS.IN_FEATURES": [
+      "p2",
+      "p3",
+      "p4",
+      "p5"
+    ],
+    "MODEL.ROI_HEADS.NAME": "StandardROIHeads",
+    "MODEL.ROI_HEADS.NUM_CLASSES": 5,
+    "MODEL.ROI_MASK_HEAD.NAME": "MaskRCNNConvUpsampleHead",
+    "MODEL.ROI_MASK_HEAD.NUM_CONV": 4,
+    "MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION": 7,
+    "MODEL.RPN.IN_FEATURES": [
+      "p2",
+      "p3",
+      "p4",
+      "p5",
+      "p6"
+    ],
+    "MODEL.RPN.POST_NMS_TOPK_TRAIN": 1000,
+    "MODEL.RPN.PRE_NMS_TOPK_TEST": 1000,
+    "MODEL.RPN.PRE_NMS_TOPK_TRAIN": 2000
+  },
+  "eos_token_id": 2,
+  "fast_qkv": false,
+  "gradient_checkpointing": false,
+  "has_relative_attention_bias": false,
+  "has_spatial_attention_bias": false,
+  "has_visual_segment_embedding": true,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Caption",
+    "1": "Footnote",
+    "2": "Formula",
+    "3": "List-item",
+    "4": "Page-footer",
+    "5": "Page-header",
+    "6": "Picture",
+    "7": "Section-header",
+    "8": "Table",
+    "9": "Text",
+    "10": "Title"
+  },
+  "image_feature_pool_shape": [
+    7,
+    7,
+    256
+  ],
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Caption": 0,
+    "Footnote": 1,
+    "Formula": 2,
+    "List-item": 3,
+    "Page-footer": 4,
+    "Page-header": 5,
+    "Picture": 6,
+    "Section-header": 7,
+    "Table": 8,
+    "Text": 9,
+    "Title": 10
+  },
+  "layer_norm_eps": 1e-05,
+  "max_2d_position_embeddings": 1024,
+  "max_position_embeddings": 514,
+  "max_rel_2d_pos": 256,
+  "max_rel_pos": 128,
+  "model_type": "layoutlmv2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "rel_2d_pos_bins": 64,
+  "rel_pos_bins": 32,
+  "shape_size": 128,
+  "tokenizer_class": "LayoutXLMTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.26.1",
+  "type_vocab_size": 1,
+  "vocab_size": 250002
+}

checkpoint-5100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cb0c893d7ba9246b2605b81283b6af8ca35042b6581a134ca20f7ef1f6823ec
+size 2927175013

checkpoint-5100/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33319f8880258e404f6ffab3d9d3d5a406eb5438df23dae3d6a8799edb36e098
+size 1476527911

checkpoint-5100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94571554363ae105b9a019e1e7654b4cc04211f994f51985bc333de223a0f640
+size 14503

checkpoint-5100/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38dfb27d63f427110c7e5cc17085bf59e6c7092ff2cc105f29c264210066e28a
+size 559

checkpoint-5100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3f663c9505650c445b2900426662d8fcee79af433cecbd1484710744afd8b6b
+size 623

checkpoint-5100/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

checkpoint-5100/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-5100/tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "cls_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "eos_token": "</s>",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "microsoft/layoutxlm-base",
+  "only_label_first_subword": true,
+  "pad_token": "<pad>",
+  "pad_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "pad_token_label": -100,
+  "sep_token": "</s>",
+  "sep_token_box": [
+    1000,
+    1000,
+    1000,
+    1000
+  ],
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": null,
+  "tokenizer_class": "LayoutXLMTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-5100/trainer_state.json

@@ -0,0 +1,280 @@
+{
+  "best_metric": 0.7223092684102689,
+  "best_model_checkpoint": "DocLayNet/layout-xlm-base-finetuned-DocLayNet-base_lines_ml384-v1/checkpoint-4200",
+  "epoch": 2.11882010801828,
+  "global_step": 5100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.8412902144953226,
+      "eval_f1": 0.13109765966400155,
+      "eval_loss": 0.5185014605522156,
+      "eval_precision": 0.14371894960965223,
+      "eval_recall": 0.1205141938939475,
+      "eval_runtime": 363.2958,
+      "eval_samples_per_second": 5.648,
+      "eval_steps_per_second": 0.355,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.372060857538036e-05,
+      "loss": 0.9231,
+      "step": 500
+    },
+    {
+      "epoch": 0.25,
+      "eval_accuracy": 0.8751371332529899,
+      "eval_f1": 0.5030902012034086,
+      "eval_loss": 0.4107707738876343,
+      "eval_precision": 0.4636655625815517,
+      "eval_recall": 0.5498422900672499,
+      "eval_runtime": 366.1838,
+      "eval_samples_per_second": 5.604,
+      "eval_steps_per_second": 0.352,
+      "step": 600
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.8887072454995973,
+      "eval_f1": 0.520614670919107,
+      "eval_loss": 0.3911038935184479,
+      "eval_precision": 0.5076044552496184,
+      "eval_recall": 0.5343093495209189,
+      "eval_runtime": 358.5079,
+      "eval_samples_per_second": 5.724,
+      "eval_steps_per_second": 0.36,
+      "step": 900
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9159741458910434e-05,
+      "loss": 0.369,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5,
+      "eval_accuracy": 0.8723944681931917,
+      "eval_f1": 0.5365372697053354,
+      "eval_loss": 0.41175082325935364,
+      "eval_precision": 0.5093875367745386,
+      "eval_recall": 0.5667440338034875,
+      "eval_runtime": 370.2231,
+      "eval_samples_per_second": 5.543,
+      "eval_steps_per_second": 0.348,
+      "step": 1200
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.7620806401969838e-05,
+      "loss": 0.2737,
+      "step": 1500
+    },
+    {
+      "epoch": 0.62,
+      "eval_accuracy": 0.8960354303687331,
+      "eval_f1": 0.6032919847328244,
+      "eval_loss": 0.33280590176582336,
+      "eval_precision": 0.6045538755752107,
+      "eval_recall": 0.6020353508302089,
+      "eval_runtime": 369.9685,
+      "eval_samples_per_second": 5.546,
+      "eval_steps_per_second": 0.349,
+      "step": 1500
+    },
+    {
+      "epoch": 0.75,
+      "eval_accuracy": 0.9185901350526106,
+      "eval_f1": 0.640386343216532,
+      "eval_loss": 0.2983764410018921,
+      "eval_precision": 0.6061765800244512,
+      "eval_recall": 0.6786883294649765,
+      "eval_runtime": 366.7561,
+      "eval_samples_per_second": 5.595,
+      "eval_steps_per_second": 0.352,
+      "step": 1800
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6084949215143123e-05,
+      "loss": 0.2542,
+      "step": 2000
+    },
+    {
+      "epoch": 0.87,
+      "eval_accuracy": 0.9162879176822183,
+      "eval_f1": 0.6593356503902912,
+      "eval_loss": 0.3114674985408783,
+      "eval_precision": 0.6323842832941691,
+      "eval_recall": 0.6886865440695114,
+      "eval_runtime": 356.8285,
+      "eval_samples_per_second": 5.751,
+      "eval_steps_per_second": 0.362,
+      "step": 2100
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.919752052270062,
+      "eval_f1": 0.6536656236030397,
+      "eval_loss": 0.2877778708934784,
+      "eval_precision": 0.6160408657643899,
+      "eval_recall": 0.6961852050229126,
+      "eval_runtime": 352.6194,
+      "eval_samples_per_second": 5.819,
+      "eval_steps_per_second": 0.366,
+      "step": 2400
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.4546014158202526e-05,
+      "loss": 0.1938,
+      "step": 2500
+    },
+    {
+      "epoch": 1.12,
+      "eval_accuracy": 0.9165257053918363,
+      "eval_f1": 0.6752131725962951,
+      "eval_loss": 0.34137091040611267,
+      "eval_precision": 0.6672865694194223,
+      "eval_recall": 0.6833303576742248,
+      "eval_runtime": 349.7664,
+      "eval_samples_per_second": 5.867,
+      "eval_steps_per_second": 0.369,
+      "step": 2700
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3007079101261929e-05,
+      "loss": 0.1581,
+      "step": 3000
+    },
+    {
+      "epoch": 1.25,
+      "eval_accuracy": 0.9193251152459752,
+      "eval_f1": 0.6871282417320962,
+      "eval_loss": 0.3610578179359436,
+      "eval_precision": 0.6867605968729564,
+      "eval_recall": 0.6874962804261144,
+      "eval_runtime": 346.075,
+      "eval_samples_per_second": 5.929,
+      "eval_steps_per_second": 0.373,
+      "step": 3000
+    },
+    {
+      "epoch": 1.37,
+      "eval_accuracy": 0.9256291916839153,
+      "eval_f1": 0.6821629855293222,
+      "eval_loss": 0.2762657403945923,
+      "eval_precision": 0.6987891648982649,
+      "eval_recall": 0.6663095875736476,
+      "eval_runtime": 342.8929,
+      "eval_samples_per_second": 5.984,
+      "eval_steps_per_second": 0.376,
+      "step": 3300
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1471221914435212e-05,
+      "loss": 0.1428,
+      "step": 3500
+    },
+    {
+      "epoch": 1.5,
+      "eval_accuracy": 0.9287285383081404,
+      "eval_f1": 0.7083662914511712,
+      "eval_loss": 0.306523859500885,
+      "eval_precision": 0.7245907761249767,
+      "eval_recall": 0.6928524668214009,
+      "eval_runtime": 342.9768,
+      "eval_samples_per_second": 5.983,
+      "eval_steps_per_second": 0.376,
+      "step": 3600
+    },
+    {
+      "epoch": 1.62,
+      "eval_accuracy": 0.9193764557741881,
+      "eval_f1": 0.6812402542881132,
+      "eval_loss": 0.29422909021377563,
+      "eval_precision": 0.6865518283469326,
+      "eval_recall": 0.6760102362673333,
+      "eval_runtime": 340.2288,
+      "eval_samples_per_second": 6.031,
+      "eval_steps_per_second": 0.379,
+      "step": 3900
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.932286857494614e-06,
+      "loss": 0.1025,
+      "step": 4000
+    },
+    {
+      "epoch": 1.74,
+      "eval_accuracy": 0.9346597203832706,
+      "eval_f1": 0.7223092684102689,
+      "eval_loss": 0.2989632189273834,
+      "eval_precision": 0.7315227342081172,
+      "eval_recall": 0.7133250014878295,
+      "eval_runtime": 368.8508,
+      "eval_samples_per_second": 5.563,
+      "eval_steps_per_second": 0.35,
+      "step": 4200
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 8.393351800554018e-06,
+      "loss": 0.1225,
+      "step": 4500
+    },
+    {
+      "epoch": 1.87,
+      "eval_accuracy": 0.9360324039797016,
+      "eval_f1": 0.7047921954799842,
+      "eval_loss": 0.2729274332523346,
+      "eval_precision": 0.7248993457473578,
+      "eval_recall": 0.6857703981431887,
+      "eval_runtime": 366.3118,
+      "eval_samples_per_second": 5.602,
+      "eval_steps_per_second": 0.352,
+      "step": 4500
+    },
+    {
+      "epoch": 1.99,
+      "eval_accuracy": 0.9395776025594604,
+      "eval_f1": 0.7221742349457059,
+      "eval_loss": 0.2825567126274109,
+      "eval_precision": 0.7496957663485557,
+      "eval_recall": 0.6966017972981016,
+      "eval_runtime": 379.1173,
+      "eval_samples_per_second": 5.413,
+      "eval_steps_per_second": 0.34,
+      "step": 4800
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.85441674361342e-06,
+      "loss": 0.108,
+      "step": 5000
+    },
+    {
+      "epoch": 2.12,
+      "eval_accuracy": 0.9301336475013375,
+      "eval_f1": 0.7193125072581582,
+      "eval_loss": 0.3070518672466278,
+      "eval_precision": 0.7022277648659373,
+      "eval_recall": 0.7372493007201095,
+      "eval_runtime": 371.7164,
+      "eval_samples_per_second": 5.52,
+      "eval_steps_per_second": 0.347,
+      "step": 5100
+    }
+  ],
+  "max_steps": 7221,
+  "num_train_epochs": 3,
+  "total_flos": 1.6514319414081024e+16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-5100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bed169afccd5558a4f44475025c0fdb96a2d7b7a0b6b87a8facddae910e56d4e
+size 3631

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8eadfa9997d1faa34349dfa531c88cf8b5dd425c80a137295d5efc012e1dcb7
+oid sha256:33319f8880258e404f6ffab3d9d3d5a406eb5438df23dae3d6a8799edb36e098
 size 1476527911

runs/Mar03_10-39-21_7e37b16db853/1677841145.0698338/events.out.tfevents.1677841145.7e37b16db853.127.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bedf6a7d3500dd27329d470546966dafca16d46f6f48435235ce5d2848646bd4
+size 5889

runs/Mar03_10-39-21_7e37b16db853/events.out.tfevents.1677841144.7e37b16db853.127.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f92a9ea0e1ec020eef85f261283bb3fe1a3ae6c245eb3689bcf8c180df3bd2f8
+size 7236

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64fe5cba1b7f9d3bec562632694d0a69501b8cea02c7e57f6ebc355e549d19b9
+oid sha256:bed169afccd5558a4f44475025c0fdb96a2d7b7a0b6b87a8facddae910e56d4e
 size 3631