Python for ML Projects | Dataford Interview Questions

Tell me about your hands-on Python experience and how you use it in ML or AI projects. Implement that experience as a Python function that converts one model's logits into its top-k class predictions. Use numerically stable softmax, temperature scaling, deterministic tie-breaking by lower class index, and probabilities rounded to six decimal places. The function accepts logits, k, and temperature, and returns [class_index, probability] pairs ordered by descending probability.

Constraints

1 <= len(logits) <= 1000
1 <= k <= len(logits)
temperature > 0
-10^9 <= logits[i] <= 10^9
Return probabilities rounded to six decimal places

Function Signature

def softmax_top_k(logits, k, temperature):

Your question is Python for ML Projects. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Constraints

1 <= len(logits) <= 1000
1 <= k <= len(logits)
temperature > 0
-10^9 <= logits[i] <= 10^9
Return probabilities rounded to six decimal places

Function Signature

def softmax_top_k(logits, k, temperature):

Your question is Python for ML Projects. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.