GeoMariano

Geographer and Lead GeoData Scientist at @HomesEngland. Daddy of 2, Raspberry Pi Geek, and enthusiastic mountaineering. Also a Spaniard ( Valencia) in the UK.

[GeoMariano @Geo_mpm]

@JaimeObregon Prueba con trigams o bigrams, si usas duckdb Como backend tiene muy buenas extensiones!

[GeoMariano @Geo_mpm]

@owenboswarva Interesting point in their guidance about data quality "The data cannot be altered once it has been lodged on the register. " Which worries if assessors have missed UPRN or not added correctly the details for addresses

[GeoMariano @Geo_mpm]

@Gaming_For_Ages @topvint Same as you, went to retro six once and never again. I went to top vint after and just the treatment, work ethics and work quality was much better. I am very happy to go back to Topvint, an honest and very reliable knowledge person !

[GeoMariano @Geo_mpm]

@DataSciNews @isaaccorley_ @calebrob6 Very early prototype stages to see how "easy" is to apply a model like unet to detect buildings as an example

[GeoMariano @Geo_mpm]

@owenboswarva Yep that is what I find strange. When we requested the lookup between council tax band and uprn in bulk the request wasn't even considered from VOA. Interesting that LA can release the information though, without any restriction from VOA at all.

[GeoMariano @Geo_mpm]

@owenboswarva I know in one of the OS products , the cross related identifier contains the reference for council tax, but I have found nearly impossible to find the council tax band. Even requesting that data to the VOA was an impossible task . So very interesting to see this council tax data

[GeoMariano @Geo_mpm]

@ndiazOk Tiene algún nombre específico para poder testearlo?

[GeoMariano @Geo_mpm]

@robhawkes I believe National Grid provides open data for free the data with all the major lines as well as towers. It would be very interesting to see how the grid flows when is overloaded https://t.co/UmA0uJs961

[GeoMariano @Geo_mpm]

@thomasforth @TheDataCity A massive credit for the lead author of splink @RobinLinacre and his team that are always releasing improvements, but also other variations open source such as a Geocoder linkage too. Civil service ideas and code for the public !

[GeoMariano @Geo_mpm]

@Autoglass After a month waiting for your repair for a small crack now I have the fully windscreen cracked. I need the car because I have to go every day to the hospital because of family health issues. Logged a complaint 10 days ago and no answer. Next step is the ombudsman 🧐

[GeoMariano @Geo_mpm]

@svpino Thanks for sharing. Another convert to uv here, simplicity and efficiency as it best

[GeoMariano @Geo_mpm]

@giswqs Apologies for having another question professor. Is the geoai package expecting rasters and geojson for instance segmentation only in 4326 CRS? Or will accept any others? I've had some issues with my local crs 27700, but couldn't find in the documents if they need to be reproject

[GeoMariano @Geo_mpm]

@giswqs dear professor, I've been following your great tutorial for Building Instance Segmentation using the GeoAI library. Would it be possible to train at the same time roads and other features as well as buildings? Or would it be better to have them separately? Many thanks

[GeoMariano @Geo_mpm]

@giswqs Super useful. I've been following the steps in the tutorial and super well explained with loads of extra information. Thank you for doing AI for geospatial much more accessible for the Geospatial community. This is a game changer with just one library .

[GeoMariano @Geo_mpm]

@giswqs Thank you very much professor. I thought that would be the case for instance segmentation. I'm training our own data based on new developments for the UK and I was interested in detecting also green areas and infrastructure areas such as roads or footpaths.

[Khuyen Tran @KhuyenTran16]

Transform document images into structured data with LlamaParse (automated validation) 📊

Converting document images such as receipts to structured spreadsheet data requires tedious typing and careful validation.

LlamaParse automates document data extraction by combining OCR parsing with schema validation, eliminating manual typing and human error.

Here is an example pipeline for extracting receipt data:
• Parse receipt images to markdown using LlamaParse OCR engine
• Define receipt structure with Pydantic models (company, date, items, totals)
• Extract structured data automatically with OpenAI integration
• Validate types and enforce business rules (positive prices, valid dates)
• Export to pandas DataFrames or spreadsheets for analysis

🚀 Full article: https://t.co/ChXhloyrsL
☕️ Run this code: https://t.co/GcBVwqlN0V

#DataScience #Python #MachineLearning #AI

[Milos @milos_gis]

I’ve been deep in R for GIS mapping for years, but today marks a thrilling pivot: my very first Python tutorial is live!

We’ll extract Google’s latest satellite embeddings, run k-means analysis, and map our results

Everyone, watch the tutorial 👉https://t.co/tAnUvyOFDI