Open
Description
How do these files work in the project? (it is a bit different from your essay)
./PROGRAM_DEMO/data/
1. data
2. dummy_data
2.1.dstc
2.2.msmarco
2.3.reddit
2.4.unifedqa
3.grounded
4.ungrounded
Also, I wonder how can I train the project with my own datasets?
I wonder if I am right to use the
EXAMPLE.JSON
to train directly. If it is wrong, how can I do it CORRECTLY?
{
"Context": "Please remind me of calling to Jessie at 2PM.",
"Knowledge": "reminder_contact_name is Jessie, reminder_time is 2PM",
"Response": "Sure, set the reminder: call to Jesse at 2PM"
},
Metadata
Assignees
Labels
No labels