RameshBal/LeanContractModel

1---2license: mit3metrics:4- accuracy5- f16---7# Model Card for Model ID8 9This model is sequential classification model that will take string content and categorize the 10content as what kind of Lean Contract is. This is a proof-of-concept and is constantly getting 11updated to improve the accuracy.12 13## Model Details14This is a research model, intended to show that we can use AI models to identify whether 15Lean principles are embedded in a contract or not 16 17### Model Description18 19This model is finetuned on Bertbasecase sequential classifier. The training data is created from a 20subject matter SMEs from Civil Engineering Domain.21 22Label Info : 23{ 'Lean_Prevents_Delay': 0,24  'Misuse_No_waiver': 1,25  'Lean_Promotes_Early_Completion': 2,26  'Lean_Prevents_Waiting' : 3,27  'Misuse_ambiguous' : 4,28  'No_Damages_Provision' : 5,29   'Non_compliance_Unaddressed' : 6,30  'No_timelines' : 7,31  'Lean_Prevents_Disputes' : 8,32 'Misuse_one_sided' : 9,33'Liability_undefined' : 10,34'Waiting' : 11,35'Lean_Prevents_Rework' : 12,36'Misuse_unfairness' : 13 }37 38 39## Getting started40 41from transformers import AutoTokenizer, BertForSequenceClassification42import torch43 44tokenizer = AutoTokenizer.from_pretrained("RameshBal/LeanContractModel")45 46model = BertForSequenceClassification.from_pretrained("RameshBal/LeanContractModel")47 48inputs = tokenizer("All disagreements between the contracting parties, with the exception of those 49explicitly denoted as finally and absolutely binding by the Contract, shall, following written 50notification by one party to the other, be resolved through arbitration. 51The arbitration process will be administered by an Engineer officer designated by the 52authority detailed in the tender documents.", return_tensors="pt")53 54with torch.no_grad():55  logits = model(**inputs).logits56 57predicted_class_id = logits.argmax().item()58 59model.config.id2label[predicted_class_id]60