At Demyst, we are constantly trying to push the boundaries to make one-click access to external data possible and reduce friction to accessing alternative data. In addition to the hundreds of external data products available through our platform and our real-time Python API, we are now introducing an enhancement to our Python package experience: Type Detection.
Through Type Detection, we infer the datatype of your input file and map it to our own internal type ontology, Demyst Types. This system considers both the naming convention and the data types of each and every data product for you, enabling you to focus on analyzing the data output. This one check process streamlines obtaining external data as our technology handles the rest of the ETL processes for all the external data products.
When you can load your input files directly, our system now automatically structures it into a dataframe and assigns it a Demyst Type by using a type guesser and convertor.
You can use this dataframe to either search for data products or run enrichments against data products. To give you easier access to the enriched data so that you are able to make quicker decisions about the product and the attributes, you can use our new sample enrichment functionality which:
- Infers the Demyst Types
- Searches for matching data products using the inferred types
- Runs an enrichment with all the searched data products using only the first thirty rows
With just a click, you are closer to an enriched dataset on your input file which can be used for sample size estimation, product selection or improving your model performance. You can utilize this in a variety of ways and try it out yourself here.