Find answers from the community

Updated 2 years ago

Hey folks when using the wikipedia

At a glance

Hey folks.. when using the wikipedia reader if I do

documents = loader.load_data(pages=['Rome', 'Paris'])
index = GPTSimpleVectorIndex(documents)
response = index.query("summarize")

it only seems to use the last document (eg Paris in this case). Am I missing something about the usage pattern

3 comments

jjerryjliu0

for summarization, i would not use GPTSimpleVectorIndex. Try using GPTListIndex instead

ffirasd

thanks.. that seemed to work better

jjerryjliu0

yeah because GPTSimpleVectorIndex will return the top k chunks, not necessarily covering the entire document

Add a reply