wenmin-wu

wenmin-wu/ds-skills

Skill Claude CodeCodex

Manual stable diffusion inference loop with classifier-free guidance that interpolates between unconditional and conditional noise predictions for controllable image generation.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Uses a binary classifier's probability to gate object detector outputs in three tiers: keep detections, append a no-finding box, or replace all detections.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Scores image aesthetic quality by passing L2-normalized CLIP ViT-L/14 embeddings through a trained MLP regressor head.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Generate descriptive text prompts from images by combining BLIP captioning with CLIP cosine similarity against curated label banks for medium, movement, and flavor attributes.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Strips the global pool and FC head from a pretrained CNN to expose spatial feature maps (H x W x C) for attention-based decoding.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Fuse CNN image features with a small tabular MLP branch via concat before a final classifier, training both branches end-to-end.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Extract per-frame CNN features then classify the temporal sequence with stacked GRU layers and a boolean mask for variable-length video inputs.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Convert per-instance RLE or polygon annotations to COCO JSON format for seamless use with Detectron2 and MMDetection.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Match two unequal-length point sets by enumerating which points to drop from the longer set, searching over deletion combinations.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Passes N tiles independently through a shared CNN backbone, concatenates their feature maps spatially, then pools for classification — a lightweight multi-instance learning approach.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Assigns multiple labels per sample using a confidence threshold on sigmoid outputs with a fallback negative class.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Extracts object centers from 3D segmentation masks using connected-component labeling and centroid computation.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Stack 1D convolutions for local feature extraction before bidirectional LSTMs to classify variable-length stroke sequences into hundreds of doodle categories.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Stratify train/validation split by binned mask coverage percentage to ensure balanced foreground representation in segmentation tasks.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Drop false-positive detections by requiring matched per-frame box counts across paired synchronized camera views (Endzone vs Sideline) — if both views see the same scene at the same instant, true events should appear in both.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Detect inverted CT slice ordering by comparing ImagePositionPatient[2] (the Z coordinate) of the first and last DICOM slice in a series, flipping the volume along axis 0 when needed so every patient ends up in canonical head→feet order.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Add a cumulative sum channel along the vertical axis to capture directional structural trends in grayscale images for segmentation.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Fuses 3D object detections from multiple models by clustering nearby predictions with DBSCAN and taking cluster centroids.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Run DeepSort on per-frame detections then overwrite each track cluster's label with the most common mapped label across the track's lifetime.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Custom Detectron2 data mapper with photometric augmentations that properly transforms images, bounding boxes, and instance masks in sync.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Partition a large dataset into N balanced shards using integer key modulo arithmetic for reproducible, class-interleaved splits across CSV files.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Dice coefficient loss for pixel-level segmentation that directly optimizes the overlap between predicted and ground-truth masks.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Fix pixel overflow artifacts in 12-bit unsigned DICOM files where values wrap around at 4096.

wenmin-wu/ds-skills

Skill Claude CodeCodex

Normalize DICOM pixel values using frequency-equalized histogram bins for globally consistent non-linear intensity mapping.