password model

Introduction

The Password Model is designed to work with Credential Digger to automatically filter false positive password discoveries. It is a text classification model based on RoBERTa and is developed by SAP OSS.

Architecture

The Password Model is a text classification model that utilizes a variant of the RoBERTa architecture. It processes text inputs in English to classify them for potential password leaks.

Training

The model is fine-tuned from CodeBERT-base-mlm on a dataset specifically curated for leak detection. Detailed information on preprocessing, evaluation, and environmental impact metrics is currently lacking.

Guide: Running Locally

To run the model locally, you need to utilize the Hugging Face Transformers library. Here is a basic setup guide:

1. Install the Transformers library:

   pip install transformers
   

2. Load the model and tokenizer:

   from transformers import AutoTokenizer, AutoModelForSequenceClassification
   
   tokenizer = AutoTokenizer.from_pretrained("SAPOSS/password-model")
   model = AutoModelForSequenceClassification.from_pretrained("SAPOSS/password-model")
   

3. Inference: Use the tokenizer to prepare inputs and the model to perform predictions.

For enhanced performance, consider using cloud GPUs from providers like AWS, Google Cloud, or Azure.

License

The model is licensed under the Apache-2.0 License.