Skip to content

Automated chatbot evaluation for generated natural language responses. Published in NAACL 2019 and powers chateval.org.

Notifications You must be signed in to change notification settings

penn-nlp/chateval

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ChatEval Evaluation

A Python package for running automated evaluations on generated natural language responses. Used in production by the ChatEval platform.

Dependencies

  • python3
  • numpy
  • nltk

Usage

  1. Run pip install chateval to install the package.
  2. Import the package using from chateval.metrics import utils.
  3. Use the method as described (e.g. utils.distinct1(responses)).

Some evaluation methods that required pre-trained word embeddings take an object as input that implements __getitem__ and __contains__. An example is given below:

class Embedding:
    def __init__(self, vectors):
        self.vectors = vectors
        self.layer1_size = self.vectors.dim
    
    def __getitem__(self, word):
        return self.vectors.query(word)
    
    def __contains__(self, word):
        return word in self.vectors
    
    def dim(self):
        return self.vectors.dim

License

MIT

About

Automated chatbot evaluation for generated natural language responses. Published in NAACL 2019 and powers chateval.org.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages