Easy

Apply a Fit/Transform Pipeline

Easy

~20 min

code completion

A pipeline is an ordered list of transformation steps. Each step exposes two operations:

fit(X) → learns parameters from X and stores them internally

transform(X) → applies the learned transformation to X

The correct pattern:

1. On training data: call fit_transform(X_train) for each step, passing the result to the next step

2. On test data: call only transform(X_test) for each step — never refit

You are given a pipeline as a list of step objects, each with .fit(X) and .transform(X) methods.

Your task:

Implement:

fit_pipeline(steps, X_train) — fits all steps sequentially, returning the final transformed training data

transform_pipeline(steps, X_test) — transforms test data through already-fitted steps

Example Tests

Single mean-centering step: training output has zero mean

Input: {"X_test":[[3]],"X_train":[[1],[3],[5]],"pipeline_spec":"mean_center_1step"}

Expected: [[0]]

Two steps applied in order: test output uses train statistics only

Input: {"X_test":[[2]],"X_train":[[0],[4]],"pipeline_spec":"mean_center_then_scale_2step"}

Expected: [[0]]

Empty pipeline returns input unchanged

Input: {"X_test":[[5]],"X_train":[[1],[2]],"pipeline_spec":"empty"}

Expected: [[5]]

You can read the full problem statement above. Create a free account to run code in the browser, submit solutions, and track your progress.