Code
Load and query
Metadata is a Parquet index; imagery is cloud-optimized. Read the index, filter what you need, then fetch only those pixels by byte range.
# pip install taco-eo import taco ds = taco.load( "hf://datasets/tacofoundation/methaneset/methaneset-emit" ) df = ds.metadata() df[df["site:country"] == "Turkmenistan"]
Examples
Explore and reproduce
Each example is a small recipe: load one dataset, do one thing, and keep the result.
Multispectral
Detect plumes in Sentinel-2
Load the finetune subset, read the plume masks and train a change-detection baseline.
Open →Hyperspectral
Detect plumes in EMIT
Stream radiance hypercubes, compare the two matched-filter products and the independent masks.
Open →Metadata
Query the catalog
Filter samples by country, date or flux with SQL over the Parquet index, before downloading imagery.
Open →Bank
Inject synthetic plumes
Combine a plume-free pretraining scene with an enhancement from the plume bank.
Open →Python
Requires taco-eo.
Other clients
| Language | Install |
|---|---|
| R | install.packages("taco", repos = "https://asterisk-labs.r-universe.dev") |
| Julia | Pkg.add("Taco") |
| JavaScript | npm install @asterisk-labs/taco |