Email of record
Verbatim transcription of the EDA and data-cleaning pre-run handover email to the workstream lead, with the Tutor on copy. Recipient names and email addresses redacted as role tokens per the personal-data redaction rule (see About me). Email substance preserved exactly. The original PDF with full headers is held in the internal record.
EDA and Data cleaning design pre-run
From: Ariel Mella <ariel.mella@gmail.com>
To: [Data-Cleaning Lead] <[anonymised]>
Cc: [Tutor] <[anonymised]>
Date: Sunday, 10 May 2026 at 17:52 UK
Subject: EDA and Data cleaning design pre-run
Hi [Data-Cleaning Lead], I have done a few pre tasks so you can proceed with the EDA and Data cleaning workstream, please see below:
1) Folder structure for working files and output created. You are the owner and I am editor. Everyone else has reading permissions only on it.
[Link: Google Drive folder for the EDA and Data Cleaning workstream]
2) I produced a pre-run design document with some of the pre-requisites and specifics I will need later on for the ML workstream, including some conventions on how we will exchange files, output, etc. and I have included some logging facilities too so I can see with run transparently.
[Link: Google Doc - pre-run design document]
3) A small script for data preparation as some special characters and identation could get lost in the word document:
[Link: 02_data_preparation.py on Google Drive]
Wish you best in this task and anything you need, clarifications, doubts or whetter you just want to discuss findings just let me know and we can talk adhoc.
We are aiming to have this task completed by Saturday May 16th so I can pick up from there and begin the ML workstream.
Thanks !!
Ariel
This page is the redacted version for tutor-facing publication. The original PDF with full email headers and recipient names is held in the internal record at 00_raw-evidence/unit-6/public/ and the contact details and identifiable information at 00_raw-evidence/unit-6/internal-only/. Cross-references the Pre-run design document (Mella, 2026p) and the cleaning script (Mella, 2026q) listed in the Evidence Index.