Initial public release
GeoIntel release gates / Compile, test, contracts and builds (push) Successful in 1m49s
GeoIntel release gates / Python and npm vulnerability policy (push) Successful in 21s
GeoIntel release gates / Production AI image, SBOM and container scan (push) Successful in 5m39s
GeoIntel release gates / Deploy exact gated revision to Unraid (push) Failing after 58m43s
GeoIntel release gates / Compile, test, contracts and builds (push) Successful in 1m49s
GeoIntel release gates / Python and npm vulnerability policy (push) Successful in 21s
GeoIntel release gates / Production AI image, SBOM and container scan (push) Successful in 5m39s
GeoIntel release gates / Deploy exact gated revision to Unraid (push) Failing after 58m43s
This commit is contained in:
@@ -0,0 +1,20 @@
|
||||
# RFC-001 — Sentinel Integration
|
||||
|
||||
## Status
|
||||
Proposed for V2.
|
||||
|
||||
## Goal
|
||||
Add Sentinel-2 workflows for NDVI, NDWI, NDBI, seasonal comparison, and raster-based change detection.
|
||||
|
||||
## Required capabilities
|
||||
|
||||
- Search imagery by area/date/cloud cover.
|
||||
- Download or stream bands.
|
||||
- Clip to project area.
|
||||
- Compute indices.
|
||||
- Store derived rasters.
|
||||
- Visualize index maps.
|
||||
- Export statistics per area.
|
||||
|
||||
## Out of scope for V1
|
||||
Live Copernicus/STAC integration should not block the V1 GeoAI foundation.
|
||||
@@ -0,0 +1,19 @@
|
||||
# RFC-002 — LiDAR Workbench
|
||||
|
||||
## Status
|
||||
Proposed for V3/V4.
|
||||
|
||||
## Goal
|
||||
Support LAS/LAZ metadata, clipping, filtering, DEM/DSM derivation, and height profile analysis using PDAL and laspy.
|
||||
|
||||
## Planned workflow
|
||||
|
||||
1. Upload or reference LAZ/LAS.
|
||||
2. Extract metadata.
|
||||
3. Clip to area.
|
||||
4. Generate DEM/DSM where feasible.
|
||||
5. Store derived raster outputs.
|
||||
6. Visualize height products on the map.
|
||||
|
||||
## Out of scope for V1
|
||||
Full point cloud visualization, large-scale tiling, and 3D rendering.
|
||||
@@ -0,0 +1,19 @@
|
||||
# RFC-003 — Training Studio
|
||||
|
||||
## Status
|
||||
Proposed for V3.
|
||||
|
||||
## Goal
|
||||
Allow users to create annotation datasets and fine-tune object detection/segmentation models.
|
||||
|
||||
## Required prerequisites
|
||||
|
||||
- Annotation Studio.
|
||||
- YOLO/COCO export.
|
||||
- Dataset versioning.
|
||||
- Model registry.
|
||||
- Training run metadata.
|
||||
- Evaluation metrics.
|
||||
|
||||
## Out of scope for V1
|
||||
Model training execution. V1 focuses on inference and QA/QC.
|
||||
@@ -0,0 +1,10 @@
|
||||
# RFC-004 — QGIS Plugin
|
||||
|
||||
## Status
|
||||
Future idea.
|
||||
|
||||
## Goal
|
||||
Allow QGIS users to send selected areas/layers to GeoIntel and pull back detections, segmentations, QA results, and exports.
|
||||
|
||||
## Rationale
|
||||
This would strongly align GeoIntel with real GIS workflows, but should only be considered after API stability.
|
||||
@@ -0,0 +1,10 @@
|
||||
# RFC-005 — MLOps and Model Registry
|
||||
|
||||
## Status
|
||||
Proposed after Training Studio.
|
||||
|
||||
## Goal
|
||||
Track model versions, metrics, datasets, training parameters, artifacts, and deployment eligibility.
|
||||
|
||||
## V1 preparation
|
||||
The V1 schema includes model registry concepts, but no full MLOps workflow is required.
|
||||
Reference in New Issue
Block a user