karma/crab-webapp
1
0
Fork 0
mirror of https://github.com/karma-riuk/crab-webapp.git synced 2025-07-04 22:08:12 +02:00
Code Issues Packages Projects Releases Wiki Activity

removed TODO.md

Browse Source
This commit is contained in:
Karma Riuk
2025-06-12 17:35:33 +02:00
parent 65878f7792
commit 0954dc135e
1 changed files with 0 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
TODO.md
View File

@ -1,49 +0,0 @@
# User stories Batch 1
## [x] As a user I want to be able to download the dataset for EITHER comment generation OR code refinment
- When asking for the dataset, there is a parameter to say whether you want to download the context or
not (the state of the repo before the PR).
- TODO: actually build the datasets
## [x] As a user I want to be able to submit my predictions for the given downloaded dataset
## [x] As a user I want to be able to see the results of the code refinment, whether the compiled and tested successfully
## [x] Add a download button to the page that gives the user a slightly more detailed version of the result, showing the bleu score for each paraphrase and the number of tests passed for code refinement (?)
## [x] As a user I want to be able to see the performance of my predictions against the benchmark (comment generation: bleu score, code refinement: # tests passed)
# Work flow
First I'll do a basic express server to serve the datasets. No login no frontend. I'll implement
Batch 1 then move on to some other features (frontend, maybe auth).
# Batch 2
## [ ] As a dev I want to be able to deploy my webapp into a container
## [x] As a user I want to have a webpage to make all the actions mentioned above
## [x] As a user I want to see what the webpage is used for (inspired from https://seart-ghs.si.usi.ch)
## [x] Remove the paths utils, it's useless in python since every path is relative to where you call the script from
# DATASET
## [x] Rebuild the dataset (is doing)
## [x] Match all the current entries in dataset.json and put the selection thing inside dataset.dont-care.json
## [x] Go through manual selection of dataset.dont-care.json to fill missing repos
## [x] The step above will already create all the archives
## [x] Signal the instances that are actually covered
## [x] When submitting for refinement, give the user a link that he can put in a text box to see the results
# Paper
## [x] Write paper