Skip to content

Commit a0b2dea

Browse files
committed
Added links to issues to propose new puzzles.
1 parent e22bd83 commit a0b2dea

File tree

2 files changed

+9
-5
lines changed

2 files changed

+9
-5
lines changed

README.md

Lines changed: 7 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,11 @@
33
This repo contains a dataset of python programming puzzles which can be used to teach and evaluate
44
an AI's programming proficiency. We hope this dataset with **grow rapidly**, and it is already diverse in
55
terms of problem difficult, domain,
6-
and algorithmic tools needed to solve the problems. Please [contribute](../../wiki/How-to-add-a-puzzle)!
6+
and algorithmic tools needed to solve the problems. Please
7+
[propose a new puzzle](/issues/new?assignees=akalai&labels=New-puzzle&template=new-puzzle.md&title=New+puzzle)
8+
or [browse the
9+
proposed puzzles](/issues?q=is%3Aopen+is%3Aissue+label%3ANew-puzzle)
10+
or [contribute through pull requests](../../wiki/How-to-add-a-puzzle).
711

812
To learn more about how well AI systems such as GPT-3 can solve these problems, read our paper:
913

@@ -79,11 +83,11 @@ and [math problems](https://en.wikipedia.org/wiki/List_of_unsolved_problems_in_m
7983
## Notebooks
8084

8185
The [notebooks](/notebooks) subdirectory has some relevant notebooks. [Intro.ipynb](/notebooks/Intro.ipynb)
82-
has a sampling of puzzles indicating which ones the AI solved and did not [Try the notebook at Binder](https://aka.ms/python_puzzles)
86+
has a dozen puzzles indicating which ones the AI solved and did not [Try the notebook at Binder](https://aka.ms/python_puzzles)
8387
and see how your programming compares to the AI baselines!
8488

8589
[Demo.ipynb](/notebooks/Demo.ipynb)
86-
has the 30 problems completed by our users in a user study. [Try the notebook at Binder](https://aka.ms/python_puzzles)
90+
has the 30 problems completed by our users in a user study. Try the [demo notebook](https://aka.ms/python_puzzles_study)
8791
and see how your programming compares to the AI baselines!
8892

8993
### Hackathon

notebooks/Intro.ipynb

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -25,7 +25,7 @@
2525
"* If you have some extra time and want to try the timed version, try the full 30-puzzle [user study notebook](https://aka.ms/python_puzzles_study) where you have 6 minutes to solve each puzzle and there is no cheating.\n",
2626
"* **Try solving some of the puzzles below** to get a sense of what the deep learning systems did and did not solve. These systems include [GPT-3](https://en.wikipedia.org/wiki/GPT-3) which \"picked up\" how to program from a huge text corpus that included source code. \n",
2727
"* **Read our [arXiv paper](https://arxiv.org/abs/2106.05784)**. The main contribution is introducing a dataset of this new type of problem to teach and evaluate computers (and humans) programming. While AI is world champions at games like chess and go, human programmers still way outperform *program synthesis* systems.\n",
28-
"* **Visit our open-source GitHub repository** https://github.com/microsoft/PythonProgrammingPuzzles to explore the full dataset and **please contribute your own new puzzles**: \n",
28+
"* **Visit our open-source GitHub repository** https://github.com/microsoft/PythonProgrammingPuzzles to explore the full dataset and **please contribute your own new puzzles**\n",
2929
"* **Share this notebook**: https://aka.ms/python_puzzles\n",
3030
"\n",
3131
"## Instructions to use the notebook below\n",
@@ -539,4 +539,4 @@
539539
},
540540
"nbformat": 4,
541541
"nbformat_minor": 5
542-
}
542+
}

0 commit comments

Comments
 (0)