Top-down program synthesis with a REPL and reinforcement learning

This is the repo for a project I did combining top-down program synthesis with reinforcement learning. The full blog post can be read here.

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
readme_mat		readme_mat
.gitignore		.gitignore
NN.py		NN.py
README.md		README.md
ShapeGen.py		ShapeGen.py
canv_utils.py		canv_utils.py
default_params_PT.json		default_params_PT.json
default_params_RL.json		default_params_RL.json
plot_utils.py		plot_utils.py
replot.py		replot.py
run_utils.py		run_utils.py
top_down.py		top_down.py
train_utils.py		train_utils.py

Provide feedback