ragas notebook with grok

[schuerstedt]

can you create a nice ragas tutorial using GROK API (free API) to run easy in Colab?

Decribe whats going on. With sample data and expected outputs.

Link to the readme!

[github-actions]

Now I have everything I need. Let me build the notebook:

Good, notebook created. Now update the README:

Done! Here's what was built:

ragas-grok-tutorial.ipynb — 23 cells, Colab-ready

What's inside:

• 🔑 Free Grok API setup (xAI console.x.ai) + local HuggingFace embeddings (no 2nd key needed)
• 🌍 Climate science sample dataset — 5 Q&A pairs each designed to trigger a different failure mode: perfect answer, noisy retrieval, hallucinated answer, missing context, irrelevant answer
• 📊 All 5 RAGAS metrics explained visually (how each is computed step by step)
• 🔢 Expected outputs shown as cell outputs so you know what to expect before running
• 🔬 Per-question score breakdown table with 🟢🟡🔴 indicators
• 💡 "Doctor's diagnostic guide" — what each low score means and how to fix it
• ⚡ Quick-start template cell for drop-in use with your own data

README updated with the new entry, linked to Colab.

---

📊 Session | premium: 1 (claude-sonnet-4.6) | free: 0 | standard: 0 | 🎟️ Copilot: 934/1500 (62%) | 🏃 CI: (no recent runs)

[schuerstedt]

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 466.5/466.5 kB 29.2 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 45.5/45.5 kB 4.4 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 87.7/87.7 kB 9.1 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 177.4/177.4 kB 14.5 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.5/2.5 MB 88.6 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 64.7/64.7 kB 6.4 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.0/8.0 MB 102.8 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 358.8/358.8 kB 32.1 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 57.1 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 503.5/503.5 kB 40.7 MB/s eta 0:00:00
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 51.0/51.0 kB 5.9 MB/s eta 0:00:00
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
google-colab 1.0.0 requires requests==2.32.4, but you have requests 2.32.5 which is incompatible.

[schuerstedt]

fix the pip error ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
google-colab 1.0.0 requires requests==2.32.4, but you have requests 2.32.5 which is incompatible.

[github-actions]

Fixed. Added !pip install -q "requests==2.32.4" after the main install to pin it back to the version google-colab requires, resolving the conflict.

---

📊 Session | premium: 1 (claude-sonnet-4.6) | free: 0 | standard: 0

[schuerstedt]

can you change it to work with an open ai api (I have a ali qwen key I want to use). Key I would paste into the notebook.

[github-actions]

Now I'll update the notebook to use the OpenAI-compatible API (Alibaba DashScope/Qwen):

Done! Here's what changed in ragas-grok-tutorial.ipynb:

• Install: langchain-xai → langchain-openai
• API key: prompts for DashScope key (sk-...) instead of xAI key
• LLM setup: ChatXAI(model="grok-3-mini") → ChatOpenAI(model="qwen-plus", base_url="https://dashscope.aliyuncs.com/compatible-mode/v1") — fully OpenAI-compatible, just swap the base URL
• Comments show qwen-turbo (faster/cheaper) and qwen-max (best) as alternatives

Get your DashScope key at dashscope.console.aliyun.com, paste it when prompted, and the notebook runs as-is.

---

📊 Session | premium: 1 (claude-sonnet-4.6) | free: 0 | standard: 0

[schuerstedt]

can you create a copy o the notebook with ali use and link it into the readme?

[github-actions]

The current ragas-grok-tutorial.ipynb was already updated to Qwen in the last session. I'll create ragas-ali-tutorial.ipynb as a properly-named copy and link it in the README.

Now add the README entry:

Done! Created ragas-ali-tutorial.ipynb — a copy of the Qwen/DashScope version — and added a new row in the README table with a Colab badge.

---

📊 Session | premium: 1 (claude-sonnet-4.6) | free: 0 | standard: 0