lucarom

Results 11 issues of lucarom

This error happens when I increase/change the number of files in the dataset while training in KPconv (ie : using a subset wont raise the error, but the full dataset...

help wanted

This feature involves creating a dataset that supports extracting samples using a moving window approach on a whole image stored locally or remotely. The goal is to provide a straightforward...

P1
effort : 3
refactoring
gdl_refactored

Draft for now Will follow soon...

## Summary Lightning’s **MLFlowLogger** can crash when logging checkpoints because MLflow apply **stricter validation to artifact paths**, rejecting **valid Python `Path` objects** and **common filename patterns** accepted by Lightning's `ModelCheckpoint`...

bug
good first issue
P2
effort : 0
documentation
dependencies

**Summary** The landing README needs to show details about the new version being active and the status of legacy branches. **Tasks** - [ ] Add a short paragraph explaining the...

P2
effort : 1
documentation

**Summary** The current documentation hosted on `readthedocs.io` still refers to the legacy version of the repository. ([https://geo-deep-learning.readthedocs.io/en/latest/](https://geo-deep-learning.readthedocs.io/en/latest/)). It should be updated to reflect the recent refactoring. The files used are...

P1
effort : 3
gdl_refactored
documentation

To support adoption and onboarding, the refactored GDL base version needs minimal but complete documentation alongside some notebook tutorials. This issue tracks the essential elements required to make the core...

P1
effort : 2
gdl_refactored
documentation

This parent issue tracks tasks related to CI/CD automation, code formatting standards, and issue tracking conventions. The goal is to ensure the project remains consistent, easy to contribute to, and...

P1
effort : 3
gdl_refactored
CI/CD

A SAR-based dataset and processing pipeline were introduced as part of recent development efforts (see: Lightning refactor #584). These changes adapted the model and configuration to support SAR data from...

enhancement
P1
effort : 3
refactoring
gdl_refactored
training
documentation

A project to extract water bodies from LiDAR-derived raster stacks (e.g., DTM, DSM, nDSM, TWI, intensity) was previously developed using an early version of the refactored GDL framework. This functionality...

enhancement
P1
inference
effort : 3
gdl_refactored
training
documentation