AliiaR's picture
Update README.md
ac5d16a
|
raw
history blame
863 Bytes
metadata
language:
  - en
pipeline_tag: conversational
tags:
  - psychology
  - dialogues
  - empathy
  - gpt2

Model description

DialoGPT finetuned on empathetic dialogues

Training data

It was trained on a large corpus of text, including some emotionally engaging datasets such as the "Facebook Empathetic Dialogues" dataset containing 25k conversations. A dataset of 25k conversations grounded in emotional situations to facilitate training and evaluating dialogue systems. You can find a dataset here.

How to use

>>> from transformers import AutoTokenizer, AutoModelForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("AliiaR/DialoGPT-medium-empathetic-dialogues")
>>> model = AutoModelForCausalLM.from_pretrained("AliiaR/DialoGPT-medium-empathetic-dialogues")