Training in progress, step 2

HW3_Data_ocr/dataset_dict.json

@@ -0,0 +1 @@
+{"splits": ["train", "test"]}

HW3_Data_ocr/test/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3117af89f2d6c9e9e97da01956422fa8db283d3db7e46ff94194677d214cc05d
+size 30523912

HW3_Data_ocr/test/dataset_info.json

@@ -0,0 +1,50 @@
+{
+  "citation": "",
+  "description": "",
+  "features": {
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "image": {
+      "feature": {
+        "feature": {
+          "feature": {
+            "dtype": "uint8",
+            "_type": "Value"
+          },
+          "_type": "Sequence"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "answer": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "question": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "words": {
+      "feature": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "boxes": {
+      "feature": {
+        "feature": {
+          "dtype": "int64",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "",
+  "license": ""
+}

HW3_Data_ocr/test/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "1d507536506aae4c",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": null
+}

HW3_Data_ocr/train/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af961ca326fcb4043dc1ca58e0cb26a477ab8e6a4765fa1d127994e1eca9f98d
+size 147849288

HW3_Data_ocr/train/dataset_info.json

@@ -0,0 +1,50 @@
+{
+  "citation": "",
+  "description": "",
+  "features": {
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "image": {
+      "feature": {
+        "feature": {
+          "feature": {
+            "dtype": "uint8",
+            "_type": "Value"
+          },
+          "_type": "Sequence"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "answer": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "question": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "words": {
+      "feature": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "boxes": {
+      "feature": {
+        "feature": {
+          "dtype": "int64",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "",
+  "license": ""
+}

HW3_Data_ocr/train/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "f3c0ef9e1d56ad66",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": null
+}

HW3_Encoded_te/data-00000-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af881cb86718deffe6643eab58e7c3850dff2d242add6a6b6364af3e8c65bf84
+size 3323016

HW3_Encoded_te/data-00001-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf75fa5420c178a961fa524f5e4a97b33d8a0f460dbfbcfb0b8e6de39653d9e6
+size 3323016

HW3_Encoded_te/data-00002-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:516f1983d209e6fb9b2c50b274fae7a8f97e8442e6bc52c9b314bcd04151fb90
+size 3323016

HW3_Encoded_te/data-00003-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87c9f0510b1da3610aae4d4c5c71595467dc2cd1d422ec249197d22842c6eb2d
+size 3323016

HW3_Encoded_te/data-00004-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84ff6acc8adef9c70330fe4cef7c70f1a9fa00d7f2c41c131f679648ac9099b9
+size 3323016

HW3_Encoded_te/data-00005-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05ad1cf48bd0cdcb1c813a39c9601bc0eb607d3fe642f3f920186d9f725657ed
+size 3323016

HW3_Encoded_te/data-00006-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d925cfb52bfd87af150510f9ef3022dcdafc561c86d18a89c4c27c963e79ea40
+size 3323016

HW3_Encoded_te/data-00007-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:607ee1170524c20056a71c0d9ac8539860f099c5af8498ae19416344b0677c14
+size 3323016

HW3_Encoded_te/data-00008-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c87318fbe84ae5d153dc3c262885ba5262cf52588ac82d18a1bde80505ad0875
+size 3323016

HW3_Encoded_te/data-00009-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0d88c1d7f0523b93a2de1398e07c9a0a23006e3bdad73a7c050aef9f63fe1fd
+size 3323016

HW3_Encoded_te/dataset_info.json

@@ -0,0 +1,60 @@
+{
+  "citation": "",
+  "description": "",
+  "features": {
+    "image": {
+      "feature": {
+        "feature": {
+          "feature": {
+            "dtype": "uint8",
+            "_type": "Value"
+          },
+          "_type": "Sequence"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "input_ids": {
+      "feature": {
+        "dtype": "int32",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "token_type_ids": {
+      "feature": {
+        "dtype": "int8",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "attention_mask": {
+      "feature": {
+        "dtype": "int8",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "bbox": {
+      "feature": {
+        "feature": {
+          "dtype": "int64",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "start_positions": {
+      "dtype": "int64",
+      "_type": "Value"
+    },
+    "end_positions": {
+      "dtype": "int64",
+      "_type": "Value"
+    }
+  },
+  "homepage": "",
+  "license": ""
+}

HW3_Encoded_te/state.json

@@ -0,0 +1,40 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00010.arrow"
+    },
+    {
+      "filename": "data-00001-of-00010.arrow"
+    },
+    {
+      "filename": "data-00002-of-00010.arrow"
+    },
+    {
+      "filename": "data-00003-of-00010.arrow"
+    },
+    {
+      "filename": "data-00004-of-00010.arrow"
+    },
+    {
+      "filename": "data-00005-of-00010.arrow"
+    },
+    {
+      "filename": "data-00006-of-00010.arrow"
+    },
+    {
+      "filename": "data-00007-of-00010.arrow"
+    },
+    {
+      "filename": "data-00008-of-00010.arrow"
+    },
+    {
+      "filename": "data-00009-of-00010.arrow"
+    }
+  ],
+  "_fingerprint": "c8ef0892137ccf76",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": null
+}

HW3_Encoded_tr/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbb896695747eaa30d35fc097b7e508b4a6064e1dba51cad8872c8ef57ba75d4
+size 157992720

HW3_Encoded_tr/data-00000-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e70f5c89fec743ea8d14d4b4fb2cf763044c3d3d0326e4ff47a5549318d3987
+size 15906312

HW3_Encoded_tr/data-00001-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aa80fb2ae9ba0d43446e1027b42ef13583b166476953627145a269107a653ec
+size 15906312

HW3_Encoded_tr/data-00002-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:227a0a0258ca59599c186448354245336d2aed341453e36da96f9082c6cb6c88
+size 15906312

HW3_Encoded_tr/data-00003-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23ecbda78acf3edcb8e732e8b2f4d01e923c51bea2646c59b99d6e290a489322
+size 15906312

HW3_Encoded_tr/data-00004-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57aa551d085119d4e0f88fd78b03ac3d49b41d722f150022c83b10666cb45e41
+size 15731568

HW3_Encoded_tr/data-00005-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ac89dc3ce17866225ed3d04059a86e3c1f01452ced73a93789332de99748218
+size 15731568

HW3_Encoded_tr/data-00006-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21d6e0810433ba242cc3822bb2384607aaff66f6bfaaa10474a1e080413c0432
+size 15731568

HW3_Encoded_tr/data-00007-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8c8a13bcd3a0c91e6e52781ddad8c67b2510d32913818431970e5bd2d38d249
+size 15731568

HW3_Encoded_tr/data-00008-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a3cc4a22a74e9b697ede75a0525987627dd04a3b9eefc89d196e726d2dd3b6a
+size 15731568

HW3_Encoded_tr/data-00009-of-00010.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f454865343904f346406866476cbd1dab775f9f6e0d497eab9482fc586c5a3
+size 15731568

HW3_Encoded_tr/dataset_info.json

@@ -0,0 +1,60 @@
+{
+  "citation": "",
+  "description": "",
+  "features": {
+    "image": {
+      "feature": {
+        "feature": {
+          "feature": {
+            "dtype": "uint8",
+            "_type": "Value"
+          },
+          "_type": "Sequence"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "input_ids": {
+      "feature": {
+        "dtype": "int32",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "token_type_ids": {
+      "feature": {
+        "dtype": "int8",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "attention_mask": {
+      "feature": {
+        "dtype": "int8",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "bbox": {
+      "feature": {
+        "feature": {
+          "dtype": "int64",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "_type": "Sequence"
+    },
+    "start_positions": {
+      "dtype": "int64",
+      "_type": "Value"
+    },
+    "end_positions": {
+      "dtype": "int64",
+      "_type": "Value"
+    }
+  },
+  "homepage": "",
+  "license": ""
+}

HW3_Encoded_tr/state.json

@@ -0,0 +1,40 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00010.arrow"
+    },
+    {
+      "filename": "data-00001-of-00010.arrow"
+    },
+    {
+      "filename": "data-00002-of-00010.arrow"
+    },
+    {
+      "filename": "data-00003-of-00010.arrow"
+    },
+    {
+      "filename": "data-00004-of-00010.arrow"
+    },
+    {
+      "filename": "data-00005-of-00010.arrow"
+    },
+    {
+      "filename": "data-00006-of-00010.arrow"
+    },
+    {
+      "filename": "data-00007-of-00010.arrow"
+    },
+    {
+      "filename": "data-00008-of-00010.arrow"
+    },
+    {
+      "filename": "data-00009-of-00010.arrow"
+    }
+  ],
+  "_fingerprint": "cb3a769983d499c2",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": null
+}

config.json

@@ -0,0 +1,132 @@
+{
+  "_name_or_path": "microsoft/layoutlmv2-base-uncased",
+  "architectures": [
+    "LayoutLMv2ForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "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
+  },
+  "fast_qkv": true,
+  "gradient_checkpointing": false,
+  "has_relative_attention_bias": true,
+  "has_spatial_attention_bias": true,
+  "has_visual_segment_embedding": true,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "image_feature_pool_shape": [
+    7,
+    7,
+    256
+  ],
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_2d_position_embeddings": 1024,
+  "max_position_embeddings": 512,
+  "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": 0,
+  "rel_2d_pos_bins": 64,
+  "rel_pos_bins": 32,
+  "shape_size": 128,
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.0",
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74fbba9ad6309d40399d40a303a1f30356f70d0e63b88b0115392547273c1c35
+size 802053752

preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "apply_ocr": true,
+  "do_resize": true,
+  "feature_extractor_type": "LayoutLMv2FeatureExtractor",
+  "image_processor_type": "LayoutLMv2ImageProcessor",
+  "ocr_lang": null,
+  "processor_class": "LayoutLMv2Processor",
+  "resample": 2,
+  "size": {
+    "height": 224,
+    "width": 224
+  },
+  "tesseract_config": ""
+}

runs/Nov08_17-11-57_c3571ebff4b6/events.out.tfevents.1699463599.c3571ebff4b6.131.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c428a425235dc67e9a89ebae2fb1b507f308e614edfa31e8b2bb50b74998fd9
+size 6269

runs/Nov08_17-15-33_c3571ebff4b6/events.out.tfevents.1699463762.c3571ebff4b6.131.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38d1ec12c42f6dda74fe5da396944e9f3073b9749f37ce26711614d98e918994
+size 88

runs/Nov08_17-20-47_c3571ebff4b6/events.out.tfevents.1699464083.c3571ebff4b6.30091.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b963c61d7cd631ee9cffda63ff2d72528c245286b9c2abdf42dbfa7c46ca4ce
+size 6683

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,79 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "cls_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "only_label_first_subword": true,
+  "pad_token": "[PAD]",
+  "pad_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "pad_token_label": -100,
+  "processor_class": "LayoutLMv2Processor",
+  "sep_token": "[SEP]",
+  "sep_token_box": [
+    1000,
+    1000,
+    1000,
+    1000
+  ],
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "LayoutLMv2Tokenizer",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4d5be9136ffa846888699ee1fdeb2386b5ea792d4184943ed799d4a737afc42
+size 4600