Skip to content
View knh4abt's full-sized avatar

Block or report knh4abt

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
knh4abt/README.md

Hi, I'm Nael

I build ML systems for the places they actually have to ship - cars, factories, clinics.

Mechatronics and Cyber Physical Systems MSc, Bosch ADAS background. Based in Stuttgart.

I care more about why a model fails than how well it scores.

  • driving-scene-segmentation - DeepLabV3+ vs SegFormer-B0 on Cityscapes. Where the two models disagree and why: truck/car confusion, the rider-vehicle boundary.
  • medvision - EfficientNet-B0 for colorectal cancer histology classification (99.6% val accuracy).
  • rul-for-nasa-jet-engines - Predictive maintenance on NASA C-MAPSS turbofan data.

Get in contact > nael.tarek.95@gmail.com LinkedIn > https://www.linkedin.com/in/nael-k-8016a711a/

Pinned Loading

  1. driving-scene-segmentation driving-scene-segmentation Public

    Semantic segmentation for autonomous driving: DeepLabV3+ vs. SegFormer-B0 on Cityscapes. 72.9% vs. 60.1% mIoU — the gap is not noise, it's truck→car confusion at 26%. PyTorch · ADAS safety analysis…

    Python

  2. medvision medvision Public

    Histology tissue classification using EfficientNet and PyTorch

    Python

  3. rul-for-nasa-jet-engines rul-for-nasa-jet-engines Public

    End-to-end RUL prediction for jet engines using NASA CMAPSS data, designed as a modular, reproducible ML pipeline with software-engineering structure.

    Python