Available question text
Given a buggy NumPy-based neural network notebook for MNIST digit classification, identify and fix issues such as softmax overflow and incorrect gradient backpropagation using the chain rule. Once debugged, tune hyperparameters to achieve greater than 80% classification accuracy.
Related questions
Continue with questions that share topics or the same question type.
- Convolution coding for 2d image
Coding
- Design Image Augmentation Function
Coding
- Find duplicate files and apply image operations
Coding · Arrays & Strings · Grids & Matrices
- Flipping an Image
Coding
Sources · 1 reference
These records explain where the question appears. A company tag or candidate report does not guarantee a future interview question.
Hello Interview · 1 record
Datology
Debug and Tune an MNIST Image Classifier
Reported Jul 24, 2026 · 1 source report
SENIOR
Source snapshot: Nov 12, 2026
View source