language form evaluation in python
Project description
Language Form Evaluation
This package includes efficient pure python implementation of the following metrics:
-
BLEU (in progress)
-
ROUGE (in progress)
-
- compared to original implementation: https://github.com/vrama91/cider
- With the same tokenizer
- error: < 1 %
- speed: + 181 %
- With different tokenizer (formeval cider use Regex from nltk)
- error: ~15 %
- speed: + 432 %
- With the same tokenizer
- compared to original implementation: https://github.com/vrama91/cider
-
SPICE (in progress)
All stats represented above are rough estimation
Dependencies
- python 3.6 +
- nltk 3.5+
Setup
pip install formeval
optional setup
python3 -c 'import nltk; nltk.download("wordnet")'
Example Data
Comparison
CIDEr/CIDEr-D (https://github.com/vrama91/cider):
- difference < 1%
- efficiency
Usage
See run_examples.py
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
formeval-1.0.2.tar.gz
(4.6 kB
view hashes)