Learn the general sequence of actions to automate an entire workflow
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.
Learn the general sequence of actions to automate an entire workflow
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.
Learn the general sequence of actions to automate an entire workflow
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
If you receive large documents with multiple sub documents (commercial invoices, BOLs etc..), your first step is to use /split
to determine which pages correspond to each document.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Once you've determined the documents that you're working with, you would then use /classify
to classify each of those documents by passing the respective pages for each document from the /split
response in the query parameter. If you haven't used entity breaks, you can classify an entire document without the need to specify certain pages.
The classifications, or labels
that you pass to this endpoint should correspond with the names of the schemas, or fields_to_extract
that you have built on your end for the respective document types that you process.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.
Now that you know what entities and document types you're working with, you can make a call to /extract
to extract a list of fields_to_extract
from a given document.