Comments (2)
Fixed by: e981b8b
from ai_planning_searching.
In retrospect, (1) was unecessary.
from ai_planning_searching.
Related Issues (19)
- select function should populate the mcts_tree. modify test_select to check that children nodes are appended.
- node 0 does not contain the correct max_rollout_reward value for Q_s_a value for child action '3' in test_backpropagate_statistics
- [p0] Set attention mask and pad token in beam generate HOT 1
- refactor select function HOT 1
- fix backpropagate_statistics HOT 1
- Add test to ensure that Node.current_token indeed only represents the current token in the sequence and not the entire token HOT 1
- Made code compatible with using multi-token prompt and doing MCTS one token at a time, i.e. a minor modification to the test in test_main_algorithm HOT 1
- code cleanup HOT 1
- [p1] The plots! The plots! Do a comparison with greedy decode and beam decode in a jupyter notebook
- Add option to use some other function f to score trajectories, as currently the MCTS code just implements beam search HOT 1
- Read the APPS paper and import the dataset into codebase HOT 3
- Fix segmentation fault due to load_dataset, optionally bypass the huggingface dataloader HOT 2
- [p2] try some form of parallelization
- decide on some standard for the format of executing python inputs HOT 2
- [p0] apps_dataset playground HOT 1
- [p1] rewrite reward_function_utils.py
- [p1] Build a dataloader that loads the APPS dataset question (prompt) with its associated unit tests, which allows shuffling, and multiple epoch loading, et cetera
- [p1] Refactor code to allow for passing in the question as the prompt to MCTS (main_algorithm) and the accompanying unit tests as the reward function (evaluate_full_paths)
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 ai_planning_searching.