Comments (3)
I think that's a good idea. I had a quick look and it seems that using SPICE requires an initial bash script to run. Since I would like this repository to be easy to install via pip, I will have a look how to best implement this.
The options I see now is to use setup.py to either run bash commands from within python or rewrite the bash script to a python script. I am leaning towards the second option because it might allow for better error handling since everything is in python.
from pycocoevalcap.
This is finally fixed in b6e7a05.
Note that my implementation automatically downloads the Stanford library (385 MB) required by SPICE the first time a SPICE evaluation is performed.
Moreover, SPICE takes a while the first time it is called, but it is sped up subsequently due to caching.
Because of these two points, however, the first run will be quite slow.
from pycocoevalcap.
On a different note, including SPICE will significantly increase the installed package size. The SPICE library is approx. 385 MB in size. So I think it might actually be better to only load this data when it is first used, instead of doing it in setup.py during pip installation.
from pycocoevalcap.
Related Issues (20)
- bleu error HOT 9
- why it is so slow to compute the meteor score? HOT 3
- the ratio of BLEU
- OSError: [Errno 22] Invalid argument HOT 1
- computing METEOR fails HOT 7
- Problem in tokenizer HOT 2
- stanford-corenlp-full-2015-12-09.zip is not accessible now HOT 6
- CIDEr score is 0 while all other metrics are normal HOT 6
- BrokenPipeError: [Errno 32] Broken pipe HOT 1
- OSError: [Errno 22] Invalid argument HOT 4
- Could not cache item for SPICE HOT 1
- Example of using pycocoevalcap WITHOUT coco data HOT 3
- mean maybe variance?
- Logical error in CIDEr ?
- What are the default values of parameters for meteor?
- meteor.py HOT 4
- is your code guaranteed? HOT 1
- [Errno 22] Invalid argument for self.meteor_p.stdin.flush() HOT 1
- [Errno 22] Invalid argument HOT 9
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pycocoevalcap.