Ollama embeddings not using base_url about scrapegraph-ai HOT 5 CLOSED

Hei @arsaboo thanks for pointing it out!
I have just committed a fix here ac099e2, now all the parameters of the config are passed to the OllamaEmbeddings class (streaming and temperature are removed since they are not compatible with embedders).

from scrapegraph-ai.

Hi we deployed a new version of the library (0.2.4) on pip remember to upgrade it, otherwise you will have the same error

from scrapegraph-ai.

Pls @arsaboo if you like the project remember to leave a star, it could help us to grow!

from scrapegraph-ai.

@VinciGit00 I am still seeing the error (even after updating).

ValueError: Error raised by inference API HTTP code: 404, {"error":"model 'mxbai-embed-large' not found, try pulling it first"}

pip show scrapegraphai
Name: scrapegraphai
Version: 0.2.4
Summary: A web scraping library based on LangChain which uses LLM and direct graph logic to create scraping pipelines.
Home-page: https://scrapegraph-ai.readthedocs.io/
Author: Marco Vinciguerra
Author-email: [email protected]
License: MIT
Location: c:\users\arsab\anaconda3\lib\site-packages
Requires: beautifulsoup4, faiss-cpu, google, graphviz, html2text, langchain, langchain-google-genai, langchain-openai, minify-html, pandas, python-dotenv, tiktoken, tqdm
Required-by:

@PeriniM It looks like you only added the base_url to the examples file.

from scrapegraph-ai.

@PeriniM base_url is still missing from examples/local_models/Docker/search_graph_docker.py , examples/local_models/Docker/script_generator_docker.py and many other files (assuming those files are responsible for initializing the model) .

from scrapegraph-ai.