Is the Documentation for Llama Extract Up-to-Date?

Question

https://github.com/run-llama/llama_extract - is the documentation up-to date? i cannot call create_agent method on the extractor. Can I pass an s3 url ? ---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[44], line 1
----> 1 extractor.create_agents() File ~/myenv/lib/python3.12/site-packages/pydantic/main.py:891, in BaseModel.__getattr__(self, item) 888 return super().__getattribute__(item) # Raises AttributeError if appropriate 889 else: 890 # this is the current error
--> 891 raise AttributeError(f'{type(self).__name__!r} object has no attribute {item!r}') AttributeError: 'LlamaExtract' object has no attribute 'create_agents'running version: Name: llama-extract
Version: 0.0.4

Logan M · Answer

https://github.com/run-llama/llama_extract?tab=readme-ov-file#%EF%B8%8F-this-project-has-been-moved-to-llamacloud-services

Logan M · Answer

https://github.com/run-llama/llama_cloud_services/blob/main/extract.md

Logan M · Answer

Yes, that is indeed out of date, following the banner at the top 👍 Moving llama-cloud related services to a single package/repo

kev · Answer

@Logan M I am getting an import error when I try to import from llama_cloud_services import LlamaExtract ---------------------------------------------------------------------------
ImportError Traceback (most recent call last)
Cell In[76], line 15 12 from PyPDF2 import PdfReader 13 from urllib.parse import urlparse
---> 15 from llama_cloud_services import LlamaExtract 16 from pydantic import BaseModel, Field 17 from typing import List, Optional File ~/myenv/lib/python3.12/site-packages/llama_cloud_services/__init__.py:2 1 from llama_cloud_services.parse import LlamaParse
----> 2 from llama_cloud_services.report import ReportClient, LlamaReport 3 from llama_cloud_services.extract import LlamaExtract, ExtractionAgent 5 __all__ = [ 6 "LlamaParse", 7 "ReportClient", (...) 10 "ExtractionAgent", 11 ] File ~/myenv/lib/python3.12/site-packages/llama_cloud_services/report/__init__.py:1
----> 1 from llama_cloud_services.report.report import ReportClient 2 from llama_cloud_services.report.base import LlamaReport 4 __all__ = ["ReportClient", "LlamaReport"] File ~/myenv/lib/python3.12/site-packages/llama_cloud_services/report/report.py:9 6 from datetime import datetime 7 from enum import Enum
----> 9 from llama_cloud.types import ( 10 ReportEventItemEventData_Progress, 11 ReportMetadata, 12 EditSuggestion, 13 ReportResponse, 14 ReportPlan, 15 ReportBlock, 16 ReportPlanBlock, 17 Report, 18 ) 20 if TYPE_CHECKING: 21 from llama_cloud_services.report.base import LlamaReport ImportError: cannot import name 'ReportEventItemEventData_Progress' from 'llama_cloud.types' (/Users/myenv/lib/python3.12/site-packages/llama_cloud/types/__init__.py)

Logan M · Answer

Hmm, pip install -U llama-cloud ?

Logan M · Answer

Might need to set a min version for that

kev · Answer

what version is stable? I am on:

kev · Answer

llama-cloud 0.1.13
llama-cloud-services 0.6.3

Logan M · Answer

The latest of both should be stable

Logan M · Answer

works fine for mehttps://colab.research.google.com/drive/1ZCG36eLVaaZGA0XIjJH1M5EN8QhygkCC?usp=sharing

Logan M · Answer

if you are in a notebook, make sure you restart it

kev · Answer

thanks, works.

kev · Answer

would we be able to pass s3 urls for extraction?

Logan M · Answer

I'm actually not sure! cc @neerajprad

neerajprad · Answer

would we be able to pass s3 urls for extraction?

Not at the moment. You'll need to upload the files. We'll soon be adding a text-mode where you should be able to read from other sources and pass the text directly for extraction.

biswaroop · Answer

What kind of files do you have on S3?

kev · Answer

cool, so i can read the pdf files from s3 and pass the text blob as input with the text-mode ?

neerajprad · Answer

Yes, that's the idea.

Find answers from the community

Is the Documentation for Llama Extract Up-to-Date?