Skip to content

Latest commit

 

History

History
50 lines (37 loc) · 818 Bytes

README.md

File metadata and controls

50 lines (37 loc) · 818 Bytes

LogPPT

Repository for the paper: Log Parsing with Prompt-based Few-shot Learning

Thanks for their contribution.

Requirements

Library

  1. Python 3.8
  2. torch
  3. transformers
  4. ...

To install all library:

pip install -r requirements.txt

We recommend to use conda environment.

conda create --name LogPPT python=3.8
conda activate LogPPT
pip install -r requirements.txt

2.2. Pre-trained models

To download the pre-trained language model:

$ cd pretrained_models/roberta-base
$ bash download.sh

III. Usage:

Run and evaluate LogPPT on Loghub-2k

conda activate LogPPT
./run_2k.sh

Run and evaluate LogPPT on Loghub-2.0

Please notice you need to download the full_dataset of Loghub-2.0 first.

conda activate LogPPT
./run_full.sh