Numeric CSV Data Cleanup

I’m working with a large collection of numeric datasets stored in CSV files and need them professionally cleaned so they’re ready for downstream reporting and analysis. The files contain only numbers—no text labels beyond the headers—yet they still suffer from common issues such as stray characters, inconsistent decimal formatting, empty rows, and duplicate records.

Your job is to open each CSV, identify and correct every inconsistency, and return a set of error-free files that preserve the original column structure. Alongside the cleaned data, include a brief change log that lists what was fixed (e.g., “removed 27 duplicate rows in sales_2023_Q1.csv”, “standardised thousand separators”, etc.) so the transformation is fully transparent.

Deliverables
• Cleaned CSV files, identical in layout to the originals
• One change log (plain text or spreadsheet) outlining all adjustments

I’m comfortable with whichever tool you prefer—Excel, LibreOffice Calc, Python (pandas), or specialised data-prep software—as long as the final output remains in CSV. Accuracy and attention to detail matter more than speed, so please only take the project if you have proven experience cleaning numeric datasets.