wenmin-wu

wenmin-wu/ds-skills

wenmin-wu/ds-skills

wenmin-wu/ds-skills

wenmin-wu/ds-skills

wenmin-wu/ds-skills

Split validation by unique anchor/query entities so no anchor appears in both train and val, preventing data leakage in pairwise m...

wenmin-wu/ds-skills

wenmin-wu/ds-skills

wenmin-wu/ds-skills

wenmin-wu/ds-skills

Three-stage packing refinement — uniform squeeze toward centroid, greedy compaction per object, then multi-directional local searc...

Converts a pretrained 2D CNN into a 3D CNN by recursively replacing Conv2d/BN2d/Pool2d layers and inflating kernel weights along t...

3D ResNet encoder extracts volumetric features, pools depth dimension, then feeds into a 2D UNet/FPN decoder for segmentation

Merges nearby predicted NER spans of the same class within a word-distance threshold into single coherent segments.

Applies albumentations augmentations to object detection data while preserving bbox-label correspondence via BboxParams.

Resample a 3D medical volume to a fixed network input shape using physical voxel spacing (dz, dy, dx), correcting the Z dimension ...

Probe an anonymized regression target by testing whether simple invertible transforms (2**y, exp(y), log(y), affine rescale) produ...

Parse sequential yes/no answers to build inclusion/exclusion sets, then apply compound boolean filters to narrow a candidate list

ArcFace angular margin loss layer for learning discriminative embeddings — used in image retrieval, product matching, and face rec...

wenmin-wu/ds-skills

Split a transaction table by a binary status flag (authorized vs. declined, paid vs. refunded) into two parallel sub-tables, then ...

Class-balanced log loss that weights each class by the inverse of its sample count, equalizing the contribution of minority and ma...

All-vs-all contrastive loss comparing every pair in a batch (N^2 pairs) with margin and compactification regularizer

Batch-compute perplexity for multiple texts using a causal LM with proper padding, shifted labels, and pad-token masking for effic...

Wrap albumentations bbox transforms in a 10-attempt retry loop that re-rolls the augmentation when all boxes get cropped out, prev...

wenmin-wu/ds-skills

Combines BCE-with-logits and soft Dice loss with configurable weights for binary and multilabel segmentation training.

Replace a pretrained ResNet's conv1 with a wider input conv to accept stacked BEV raster channels (semantic map + agent history) w...

Two-stage CT classifier where a 2D CNN dumps per-slice features once, then a bidirectional GRU runs over the slice sequence to pro...

wenmin-wu/ds-skills