distilbert-complaints-product

This model was trained from the CFBP dataset, also made available on the HuggingFace Datasets library. This model predicts the type of financial complaint based on the text provided

Model description

A DistilBert Text Classification Model, with 18 possible classes to determine the nature of a financial customer complaint.

Intended uses & limitations

This model is used as part of.a demonstration for E2E Machine Learning Projects focused on Contact Centre Automation:

• Infrastructure: Terraform
• ML Ops: HuggingFace (Datasets, Hub, Transformers)
• Ml Explainability: SHAP
• Cloud: AWS
  • Model Hosting: Lambda
  • DB Backend: DynamoDB
  • Orchestration: Step-Functions
  • UI Hosting: EC2
  • Routing: API Gateway
• UI: Budibase

Training and evaluation data

consumer_complaints dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3

Framework versions

• Transformers 4.16.1
• Pytorch 1.10.0+cu111
• Datasets 1.18.2
• Tokenizers 0.11.0