Training in progress, epoch 1

README.md

@@ -0,0 +1,57 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: distilbert/distilbert-base-uncased
+tags:
+- generated_from_trainer
+model-index:
+- name: ml_document_classification
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# ml_document_classification
+
+This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on the None dataset.
+It achieves the following results on the evaluation set:
+- eval_loss: 1.1420
+- eval_accuracy: 0.7438
+- eval_runtime: 3.0534
+- eval_samples_per_second: 157.204
+- eval_steps_per_second: 9.825
+- epoch: 1.0
+- step: 120
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-06
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 20
+
+### Framework versions
+
+- Transformers 4.47.1
+- Pytorch 2.5.1+cu121
+- Datasets 3.2.0
+- Tokenizers 0.21.0

config.json

@@ -0,0 +1,217 @@
+{
+  "_name_or_path": "distilbert/distilbert-base-uncased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": "3rd Party Credit Report Confirmation",
+    "1": "ACH Form & Voided Check",
+    "2": "AOO",
+    "3": "AVE",
+    "4": "Amortization Schedule",
+    "5": "Appraisal Delivery",
+    "6": "Appraisal Invoice",
+    "7": "Appraisal Review",
+    "8": "Appraisal",
+    "9": "Background Check",
+    "10": "Bank Statement",
+    "11": "Bank",
+    "12": "BankBrokerage Verification",
+    "13": "Budget Sheet",
+    "14": "Certificate of Formation",
+    "15": "Certificate of Occupancy",
+    "16": "Citizenship Afidavit",
+    "17": "Closing Info Sheet",
+    "18": "Closing Protection Letter",
+    "19": "Cogs",
+    "20": "Condo Documentation",
+    "21": "Corporate Bylaws",
+    "22": "Credit Check Authorization",
+    "23": "Credit Report",
+    "24": "DHCR Rent Regulation Report",
+    "25": "DSCR Calculation",
+    "26": "Deep Dive Report Card",
+    "27": "Driver License",
+    "28": "EDR",
+    "29": "EIN",
+    "30": "ESR",
+    "31": "Entity Operating Agreement",
+    "32": "Escrow Release Confirmation",
+    "33": "Executed Closing Package",
+    "34": "Experience Validation",
+    "35": "Extension Agreement",
+    "36": "Feasibility Invoice",
+    "37": "Feasibility Report",
+    "38": "Filing Receipt",
+    "39": "Flood Certificate",
+    "40": "Flood Insurance",
+    "41": "Funding Shield Cert",
+    "42": "General Liability Insurance",
+    "43": "Google Searches",
+    "44": "Government Issued ID",
+    "45": "HO6 Insurance",
+    "46": "HOA Dues",
+    "47": "Housing History",
+    "48": "ICE Valuation",
+    "49": "Insurance Paid Reciept",
+    "50": "Invoices",
+    "51": "Leases",
+    "52": "Letter of Explanation (LOE)",
+    "53": "Loan Application Form",
+    "54": "Loan Approvals",
+    "55": "Loan Shield",
+    "56": "Loan Sizer",
+    "57": "Mortgage Commitment - CTC",
+    "58": "Mortgage Statements",
+    "59": "Original HUD",
+    "60": "PCA",
+    "61": "Passport",
+    "62": "Payoff Statement",
+    "63": "Permits",
+    "64": "Personal Financial Statement",
+    "65": "Phase 1",
+    "66": "Pictures",
+    "67": "Plans",
+    "68": "Pre-Closing Approved HUD",
+    "69": "Property Financials",
+    "70": "Property Insurance",
+    "71": "Property Management Questionnaire",
+    "72": "Public Records Property Record & Building Department Search",
+    "73": "Purchase Contract",
+    "74": "Purchase Documents",
+    "75": "REO",
+    "76": "Refinance Disclosure Form",
+    "77": "Rent Roll",
+    "78": "Resolution",
+    "79": "Scope of Work",
+    "80": "Signed Term Sheet",
+    "81": "Site X",
+    "82": "Social Security Verification",
+    "83": "Stock Ledger",
+    "84": "T-12",
+    "85": "Tax Bill",
+    "86": "Term Sheet",
+    "87": "Title Commitment",
+    "88": "Title E&O Policy",
+    "89": "Trailing Documents",
+    "90": "TruePic",
+    "91": "USPS Verification",
+    "92": "VOM",
+    "93": "Wire Instructions"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "3rd Party Credit Report Confirmation": 0,
+    "ACH Form & Voided Check": 1,
+    "AOO": 2,
+    "AVE": 3,
+    "Amortization Schedule": 4,
+    "Appraisal": 8,
+    "Appraisal Delivery": 5,
+    "Appraisal Invoice": 6,
+    "Appraisal Review": 7,
+    "Background Check": 9,
+    "Bank": 11,
+    "Bank Statement": 10,
+    "BankBrokerage Verification": 12,
+    "Budget Sheet": 13,
+    "Certificate of Formation": 14,
+    "Certificate of Occupancy": 15,
+    "Citizenship Afidavit": 16,
+    "Closing Info Sheet": 17,
+    "Closing Protection Letter": 18,
+    "Cogs": 19,
+    "Condo Documentation": 20,
+    "Corporate Bylaws": 21,
+    "Credit Check Authorization": 22,
+    "Credit Report": 23,
+    "DHCR Rent Regulation Report": 24,
+    "DSCR Calculation": 25,
+    "Deep Dive Report Card": 26,
+    "Driver License": 27,
+    "EDR": 28,
+    "EIN": 29,
+    "ESR": 30,
+    "Entity Operating Agreement": 31,
+    "Escrow Release Confirmation": 32,
+    "Executed Closing Package": 33,
+    "Experience Validation": 34,
+    "Extension Agreement": 35,
+    "Feasibility Invoice": 36,
+    "Feasibility Report": 37,
+    "Filing Receipt": 38,
+    "Flood Certificate": 39,
+    "Flood Insurance": 40,
+    "Funding Shield Cert": 41,
+    "General Liability Insurance": 42,
+    "Google Searches": 43,
+    "Government Issued ID": 44,
+    "HO6 Insurance": 45,
+    "HOA Dues": 46,
+    "Housing History": 47,
+    "ICE Valuation": 48,
+    "Insurance Paid Reciept": 49,
+    "Invoices": 50,
+    "Leases": 51,
+    "Letter of Explanation (LOE)": 52,
+    "Loan Application Form": 53,
+    "Loan Approvals": 54,
+    "Loan Shield": 55,
+    "Loan Sizer": 56,
+    "Mortgage Commitment - CTC": 57,
+    "Mortgage Statements": 58,
+    "Original HUD": 59,
+    "PCA": 60,
+    "Passport": 61,
+    "Payoff Statement": 62,
+    "Permits": 63,
+    "Personal Financial Statement": 64,
+    "Phase 1": 65,
+    "Pictures": 66,
+    "Plans": 67,
+    "Pre-Closing Approved HUD": 68,
+    "Property Financials": 69,
+    "Property Insurance": 70,
+    "Property Management Questionnaire": 71,
+    "Public Records Property Record & Building Department Search": 72,
+    "Purchase Contract": 73,
+    "Purchase Documents": 74,
+    "REO": 75,
+    "Refinance Disclosure Form": 76,
+    "Rent Roll": 77,
+    "Resolution": 78,
+    "Scope of Work": 79,
+    "Signed Term Sheet": 80,
+    "Site X": 81,
+    "Social Security Verification": 82,
+    "Stock Ledger": 83,
+    "T-12": 84,
+    "Tax Bill": 85,
+    "Term Sheet": 86,
+    "Title Commitment": 87,
+    "Title E&O Policy": 88,
+    "Trailing Documents": 89,
+    "TruePic": 90,
+    "USPS Verification": 91,
+    "VOM": 92,
+    "Wire Instructions": 93
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.1",
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:675f0ec62015b6911edf1254a9a96e2884df17821d27f2fc928b7d6dee1203be
+size 268115568

runs/Jan12_22-17-45_6b605e9ad474/events.out.tfevents.1736720268.6b605e9ad474.3152.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1db219b67446608edfac28f79987be0d1996af295cbded3c69ef72da4c0729da
+size 14455

runs/Jan12_22-26-01_6b605e9ad474/events.out.tfevents.1736720762.6b605e9ad474.3152.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3cd85f0cebae08c8a52b13531dc5b5b58ba41a746b4affbe5fcaddb0aa0cd69
+size 18158

runs/Jan12_22-39-15_6b605e9ad474/events.out.tfevents.1736721556.6b605e9ad474.3152.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b39c6ebd99e68d1f451043e7750e0b42d499919c1eaf89741d31644cb93769a
+size 18158

runs/Jan12_22-48-41_6b605e9ad474/events.out.tfevents.1736722122.6b605e9ad474.3152.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecd2469032a5415c009a33981021d0192083c836b43802913c32093857c24a6f
+size 18158

runs/Jan12_22-57-57_6b605e9ad474/events.out.tfevents.1736722678.6b605e9ad474.3152.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7668722d4585f40aad92ae262641a4a91a7ab6c6cc9447f7418a9338025bcb19
+size 18161

runs/Jan12_23-12-43_6b605e9ad474/events.out.tfevents.1736723564.6b605e9ad474.3152.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9f1a8c38d16b9e5606d505d995602d6548881c8804e06bd590293310f8a1f9a
+size 18161

runs/Jan12_23-23-33_6b605e9ad474/events.out.tfevents.1736724214.6b605e9ad474.3152.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d9e6acbe385bbe539d6648c8a83e4e6e51adac53f99fc9f129111316458dee5
+size 18161

runs/Jan12_23-32-29_6b605e9ad474/events.out.tfevents.1736724750.6b605e9ad474.3152.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd6ae154cc2bb1ededb117a6c21c7f5a415deb3b62f7e633694b96ab35b3eb77
+size 15658

runs/Jan12_23-38-53_6b605e9ad474/events.out.tfevents.1736725133.6b605e9ad474.3152.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf28f1a7844069bc7d853fe3b612f4231e4a06f699a4113b4f0d6bf8e582e938
+size 12328

runs/Jan13_00-01-53_6b605e9ad474/events.out.tfevents.1736726514.6b605e9ad474.3152.9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:852daeff8402386fd125f7bad5676dfc95636fe2eaf591668f5e34008ef7b91e
+size 10825

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "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
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96bf5456b513b4bc5ba560a8f9477154f0dcbb80eea430a65c0b335fa0d3167d
+size 5368