Run Large Language Models Locally With *ollama.ai's* Docker Image

ollama is an open-source framework for self-hosting large language models (LLM) similar to ChatGPT or Google’s Bard. The official Docker image makes it painless to host any number of the supported models.

Shell script to get an Ollama model running #

The example “run_ollama_container.sh” script(s) below take the LLM model name as an argument (assigned to $LLMMODEL), but the pull and run docker exec commands can be called additional times to start other models as well within the same running container.

Quick and dirty geographic longitude offset for GIS

Convert GIS data with Lat/Lon (EPSG:4326/WGS84) coordinates with longitude in the range 0-360 to -360-0 to cover -180-180 in a GIS envrionments map with this quick and dirty custom proj-style projection string.

The custom projection definition is +proj=longlat +ellps=WGS84 +towgs84=0,0,0,0,0,0,0 +pm=-360 +no_defs

This can be used as follows,

1. load a spatial dataset with Lons defined in the 0-360 range,
2. duplicate or load another layer with the same dataset,
3. set layer in 1. CRS to WGS84 or EPSG4326,
4. and set layer in 2. to a custom CRS using the projection definition above.

This process results in the dataset covering the -360 to 360 Longitude range (and implcitly -180 to 180 as well) between the two layers which can be used in a variety of different project coordinate reference systems or projections. Switching +pm=-360 to +pm=+360 will go the opposite way from (-180-180) Lons to (0-360) if necessary, because that parameter sets or modifies the default Prime Meridian Longitude in degrees-East.

NWS13: NetCDF Replacement for NWS12 Met Inputs and Application

2020 ADCIRC User’s Group Meeting Slides