-
|
I ran this code and it was fine. But when I try it with
it is giving me |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 1 reply
-
|
Hello! As you can see in the Reader docs, this node supports models trained for Extractive Question Answering such as deepset/roberta-base-squad2. google/flan-t5-base is a different type of model, trained to generate text. You can use it in the PromptNode, to perform several tasks depending on the prompt, including Generative Question Answering. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you so much for you answer. Just one more thing, if I want to evaluate the model google/flan-t5-base I can't use evaluating reader. What are the other approaches I can consider |
Beta Was this translation helpful? Give feedback.
-
|
The problem of evaluating Generative Question Answering is open and complex, but very relevant. Currently, Haystack does not provide an out-of-box solution for this task, but we are working on it... Keep following the project and you will see that we will get there. 🙂 |
Beta Was this translation helpful? Give feedback.
Hello!
As you can see in the Reader docs, this node supports models trained for Extractive Question Answering such as deepset/roberta-base-squad2.
google/flan-t5-base is a different type of model, trained to generate text. You can use it in the PromptNode, to perform several tasks depending on the prompt, including Generative Question Answering.