So you want to know the scientific Python stack

The usual suspects

• NumPy for arrays and numerical work.
• pandas for tables and tabular measurements.
• matplotlib for plots and quick visual inspection.
• scikit-image for image loading, filtering, and segmentation utilities.
• scikit-learn for machine learning and model evaluation.
• SciPy for numerical methods and image-adjacent utilities.
• napari for interactive image viewing.

What each library tends to feel like in practice

NumPy: arrays everywhere

import numpy as np

image = np.array([[1, 2], [3, 4]])
print(image.shape)
print(image.mean())

pandas: tables with labels

import pandas as pd

data = pd.DataFrame(
    {
        "sample": ["wt", "mutant"],
        "cell_count": [120, 95],
    }
)
print(data)

matplotlib: just plot the thing

import matplotlib.pyplot as plt

plt.plot([1, 2, 3], [2, 5, 4])
plt.title("Quick plot")
plt.show()

Combined example

This is the kind of short mixed example that appears all over scientific Python:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

measurements = pd.DataFrame(
    {
        "cell_id": ["c1", "c2", "c3"],
        "mean_intensity": np.array([120, 151, 118]),
    }
)

measurements.plot(x="cell_id", y="mean_intensity", kind="bar", legend=False)
plt.ylabel("Mean intensity")
plt.tight_layout()

Try this for yourself

1. Import NumPy and create a small array.
2. Create a tiny pandas table with two columns.
3. Plot one of those columns with matplotlib.
4. Write one sentence on which library felt most intuitive and why.