[DOC]: Unable to find Figure 9.2 Ray Server UI as shown on P.364

[ksparkKJE]

Describe the bug

Hi benman1,
Thanks for the great book and resources!

This question is related to Figure 9.2 on page 364.
I ran the serve_index.py file as instructed, and while the server starts successfully, I can't seem to find the Ray Server UI shown in Figure 9.2.
It looks like the console screen, but the execution screen is different.

How can I access that screen?

For reference, I'm using the latest Ray version installed with the following command:
pip install "ray[all]"

Name: ray
Version: 2.48.0

Thanks in advance for your help!
Best regards,

Steps to reproduce

No response

Expected behavior

No response

Actual behavior

No response

Branch

None

Branch name

No response

Operating System

No response

Dependency management

No response

Python version

No response

LangChain version

No response

LLM Used

No response

Model used

No response

Additional context

No response

Additional context

No response

[benman1]

Hi @ksparkKJE. Thanks for the feedback and for reporting this new issue.

I've just had a look and found issues I had to fix, both in serve_index. There was one about loading the pickled index, one about the model that's used, and finally the server was shutting down - so I had to put in a loop. I've pushed the changes to fix this.

My output looks like this.

I don't know why it looks differently from before. I am assuming there were some changes in ray. In retrospect, I should have pinned the ray version as well.

The most important thing is though: can you use the API? With the example query here:
http://localhost:8000/?query=How%20can%20Ray%20help%20with%20deploying%20LLMs%3
I am getting this:

{"results":[{"content":"discovery, demonstrate the advantages of LLMs augmented with expert tools in specialized domains.\\n\\n5. **Challenges and Limitations**: The documents highlight challenges such as hallucination in model outputs and the finite context length of LLMs, which affects their ability to incorporate historical information and perform self-reflection. Techniques like Chain of Hindsight and Algorithm Distillation are discussed to enhance model performance through iterative learning.\\n\\n6. **Structured","source":"https://python.langchain.com/docs/tutorials/summarization/","score":1.1148607730865479},{"content":"<h3 class=\"anchor anchorWithStickyNavbar_LWe7\" id=\"evaluation\">Evaluation<a href=\"#evaluation\" class=\"hash-link\" aria-label=\"Direct link to Evaluation\" title=\"Direct link to Evaluation\">\u201a\u00c4\u00e3</a></h3>\n<p>LangSmith helps you evaluate the performance of your LLM applications. The tutorial below is a great way to get started:</p>\n<ul>\n<li><a href=\"https://docs.smith.langchain.com/tutorials/Developers/evaluation\" target=\"_blank\" rel=\"noopener noreferrer\">Evaluate your LLM application</a></li>","source":"https://python.langchain.com/docs/tutorials/","score":1.2119290828704834},{"content":"39086, &#x27;section&#x27;: &#x27;end&#x27;}, page_content=&#x27;}\\n]\\nChallenges#\\nAfter going through key ideas and demos of building LLM-centered agents, I start to see a couple common limitations:&#x27;), Document(id=&#x27;ca7f06e4-2c2e-4788-9a81-2418d82213d9&#x27;, metadata={&#x27;source&#x27;: &#x27;https://lilianweng.github.io/posts/2023-06-23-agent/&#x27;, &#x27;start_index&#x27;: 32942, &#x27;section&#x27;: &#x27;end&#x27;}, page_content=&#x27;}\\n]\\nThen after these clarification, the","source":"https://python.langchain.com/docs/tutorials/rag/","score":1.253100037574768},{"content":"is Generative \uf8ff\u00fc\u00a7\u00d8</p><p>LLMs are generative models, so they can do some pretty cool things like correctly extract the height of the person in meters","source":"https://python.langchain.com/docs/tutorials/extraction/","score":1.2613029479980469},{"content":"<p>In order for an LLM to be able to generate a Cypher statement, it needs information about the graph schema. When you instantiate a graph object, it retrieves the information about the graph schema. If you later make any changes to the graph, you can run the <code>refresh_schema</code> method to refresh the schema information.</p>","source":"https://python.langchain.com/docs/tutorials/graph/","score":1.29395592212677}],"status":"success","message":"Found 5 results"}

Finally, I've also just found that the performance dashboard needs to be enabled through an additional parameter include_dashboard=True in the ray.init() execution (just pushed as well). I needed to run pip install -U "ray[default]" for this. My ray version is now 2.49.0.

On localhost:8265:

I didn't exactly answer your question. I hope this helps :). Let me know, please!

[benman1]

@ksparkKJE Did this answer your question? I'm resolving this now.