Nicolò Lucchesi

`CropAndPad` interpolation doesn't default to `INTER_NEAREST` for masks

1

## 🐛 Bug Interpolation mode for masks should default to `INTER_NEAREST` when CropAndPad is applied to masks. ## To Reproduce ``` t = A.CropAndPad(percent=(-0.25, 0.25), p=1) x = np.random.randn(80, 80).astype(np.uint8)...

Stream-line of tasks with single Env

5

## Motivation and Context In domains like Continual/Lifelong Learning (or any mult-task context) it would be convenient to be able to serve multiple tasks without the need to change Env,...

opt/conda/bin/conda not found when building Dockerfile

1

## 🚀 Feature As title, I run into this error when building from Dockerfile; apparently it's caused by curl missing the "follow redirects" flag and thus not installing conda. Solution...

Functions on collections of input

12

Hey, I was wondering whether there's a more efficient way to run functions on a list of inputs (e.g. hex ids), using a simple loop for functions like `[h3.h3_to_geo_boundary(h, geo_json=True))...

ImportError: cannot import name 'Mapping' from 'collections'

3

Hey, thanks a lot for sharing this great accomplishment with the community! I have just tried running the `cli_chat` on Python3.11 and I get `ImportError: cannot import name 'Mapping' from...

Hey, I've found this crate quite useful in its simplicity, thanks for that! This PR addresses #9, providing the methods to convert from Array->Image without copies. I've pushed the first...

The repo doesn't look too pretty atm, we could use some Ops

Implement PPO algorithm from scratch following the framework guidelines so that we can easily integrate it with all available features.

Current module structure is borrowed from avalanche but file location for rl utils isn't super-obvious. I believe we should move stuff around in order to make it more predictable for...