Data samples can be ingested to Substra using our python SDK or our command line interface tool.
2 different modes are available when adding data samples:
- local: data are located on the local filesystem and are transferred through HTTP requests. This mode is most likely to be used to add relatively small files or for testing purposes.
- remote: data are located on the server filesystem. Data must have been transferred previously to the server. Please ask the substra server administrator for further details regarding how to transfer the data. This mode is recommended for adding relatively big data samples or for adding many data samples.