The Internora

Internship field

Data Science & Analytics

A hands-on analytics track. You work with real datasets from the first week: clean them, describe them, visualise them, then model them and write the findings up for a non-technical reader.

PythonPandasNumPyData cleaningStatisticsMatplotlibEDAscikit-learnReporting
Program length:4 weeks · PKR 700

Week 1 — Python for data

The Python you actually need for analysis.

Objectives

  • Work with lists, dicts and comprehensions
  • Read files into Python
  • Use NumPy arrays

Tasks (3)

  1. 1. Python basics set

    Solve eight short problems covering loops, functions, dictionaries and comprehensions.

    Deliverable: Notebook (.ipynb) or .py file.

    10 pts

  2. 2. Read and summarise a file

    Load a CSV without Pandas, count rows, find unique values in one column and compute a mean.

    Deliverable: Notebook with output.

    10 pts

  3. 3. NumPy and Pandas intro

    Load the same CSV with Pandas, then use head, info, describe, value_counts and basic indexing.

    Deliverable: Notebook with output.

    15 pts

Week 2 — Data cleaning

Turn a messy dataset into an analysable one.

Objectives

  • Handle missing values
  • Fix types and duplicates
  • Document every cleaning decision

Tasks (3)

  1. 1. Missing values

    Profile missing values in the provided dataset and handle each column with a justified strategy.

    Deliverable: Notebook plus a short rationale.

    15 pts

  2. 2. Types, duplicates and outliers

    Fix data types, remove duplicates, standardise text categories and flag outliers.

    Deliverable: Notebook plus the cleaned CSV.

    20 pts

  3. 3. Cleaning log

    Write a cleaning log listing every change, the reason and how many rows it affected.

    Deliverable: PDF or markdown.

    10 pts

Week 3 — Analysis

Describe the data and find what is actually going on.

Objectives

  • Compute descriptive statistics
  • Group and aggregate
  • Identify trends

Tasks (3)

  1. 1. Descriptive statistics

    Compute mean, median, mode, standard deviation and quartiles for every numeric column, and interpret two of them.

    Deliverable: Notebook with interpretation.

    10 pts

  2. 2. Group analysis

    Use groupby to compare at least three categories on two metrics and explain the difference.

    Deliverable: Notebook with output.

    15 pts

  3. 3. Trend analysis

    Analyse a time or ordered dimension and describe the trend in plain language.

    Deliverable: Notebook plus a written finding.

    15 pts

Week 4 — Visualisation

Final project

Make charts that carry the finding, not just the data.

Objectives

  • Choose the right chart type
  • Label charts properly
  • Build a small dashboard

Tasks (3)

  1. 1. Chart set

    Produce six charts: distribution, comparison, composition, relationship, trend and ranking. Every chart needs a title, axis labels and units.

    Deliverable: Notebook plus exported images.

    20 pts

  2. 2. Chart critique

    Pick your two weakest charts, explain what is wrong and submit improved versions.

    Deliverable: Before and after images plus notes.

    10 pts

  3. 3. Mini dashboard

    Combine four charts into a single dashboard layout with a short summary panel.

    Deliverable: Image or notebook export.

    20 pts