edfEazyDataFix

Assessment

edf.assess() produces a QualityReport — a structured, serialisable snapshot of everything EazyDataFix knows about your dataset.

What is assessed

  • Row and column counts
  • Missing values (per column and overall rate)
  • Duplicate rows (exact match)
  • Dtype consistency (values that don’t match the declared type)
  • Cardinality per column
  • Composite quality score (0–100)

Running an assessment

assessment.py
import eazydatafix as edf

report = edf.assess("employees.csv")
report.summary()
Python 3.11
>>> report.summary()
QualityReport(employees.csv)
  rows           1,204
  columns           12
  quality_score   94.0

Custom thresholds

Override the default warning levels for your domain:

thresholds.py
report = edf.assess(
    "sales.xlsx",
    thresholds={"missing": 0.02, "duplicates": 0.01},
    verbose=True,
)

Reading the report

report.summary() prints a REPL-friendly overview. report.to_dict() returns a nested dict suitable for JSON logging. report.to_html() writes an HTML report for CI artefacts.

Next

Once you know what’s wrong, use edf.fix() to apply the cleaning pipeline.