Images can be loaded from a URL or base64 data URI via any of the data formatsthat support keyed inputs or from a directory of files. See the details onfile loaders for how to load images from a directory and the image and image_manual docs for details on the expected JSONformat.
Audio and video data can be loaded from a URL or base64 data URI via any of thedata formats that support keyed inputs or from a directory of files. See thedetails on file loaders for how to load images from a directoryand the audio and audio_manual docs for details on theexpected JSON format.
Out-of-the-box, Prodigy currently supports loading in data from single files ofJSONL, JSON, CSV or plain text. You can specify the loader via the --loaderargument on the command line. If no loader is set, Prodigy will use the fileextension to pick the respective loader. Loaders are available viaprodigy.components.loaders.
In your case, you already have the loader and CustomCsvLoader(source) is the loaded source, i.e. the data stream. So you can pass this in as the source argument, or do result['stream'] = CustomCsvLoader(source).
df19127ead