This repo includes pytorch implementation of ECT-SKIE. ECT-SKIE can automatically extract relevant information from earnings call transcripts. Our model leverages the structural information in transcripts to extract key insights effectively while providing concise explanations for each decision made by the model. We hope our research can shed light on the development of more efficient and effective transcript representation learning models for financial analysis.
Download and install the environment from the requirments file.
/yourpath/anaconda3/envs/env_name/bin/python3.8 -m pip install -r requirements.txt
conda activate env_name
See main.py for possible arguments.
use ECT-SKIE to generate results to explain the key insights selection for earnings call transcripts:
python main.py --test True
DeepVIB Repo: pytorch implementation of deep variational information bottleneck.