CourseNatural Language Processing: The Complete Course · Module 8: Extraction, Question Answering and Summarisation · part 57 of 73
Part 57 · Module 8: Extraction, Question Answering and Summarisation

Lab: Booking records, help answers and thread summaries

134 min read·14 Oct 2026
Module 8 · Lab · About 60 minutes · Needs: lessons 1 to 3 of this module, and the Module 7 lab (or its cells, pasted here). CPU only; no key needed

What you are building

FOR DEV FOR INES FOR ROSA extract.py booking records qa.py answer finder summarise.py thread summaries per field customer P 0.744 EM and F1 F1 0.571 of 40 ROUGE, support 131 of 136 backed one report: reports/extract_qa_summaries.json
Three settled files (blue), one for each of Dev, Ines and Rosa, each with its own score (green), saved together in one report.

It is Thursday 29 October 2026, project week 8. Three people asked for three things this week, and each lesson built one in a notebook. Dev wants every repair booking as a checked record. Ines wants the exact words from the help centre that answer a customer's question, or nothing when the help centre has no answer. Rosa wants every long thread in 60 words she can trust.

This lab settles the three lessons' files once, as inbox/extract.py, inbox/qa.py and inbox/summarise.py with their tests, then gives each piece its own score on the shared data: field by field for Dev, exact match and F1 for Ines, ROUGE and two faithfulness checks for Rosa. It ends with one report in reports/.

Open a new Colab notebook, call it module-08-lab.ipynb, and run each cell in order.

The rest of this part opens with an account

Signing up is free and takes one email. Nothing to pay, and nothing renews.

Start reading free