Computer Vision
Build computer vision models: image classification, object detection, segmentation, data augmentation strategies, and evaluation. Covers CNN architectures, transfer learning from pretrained models, and handling small datasets. Use when working with image data.
Install
$ agentstack add skill-ericwang915-data-scientist-skills-computer-vision ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-ericwang915-data-scientist-skills-computer-vision) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Computer Vision
Purpose
Build and train computer vision models for image classification, detection, and segmentation.
How It Works
Task Selection
| Task | Architecture | Library | |------|-------------|---------| | Image classification | ResNet, EfficientNet, ViT | torchvision, timm | | Object detection | YOLOv8, Faster R-CNN, DETR | ultralytics, detectron2 | | Segmentation | U-Net, Mask R-CNN, SAM | segmentation_models | | Similarity search | CLIP, embeddings | transformers |
Data Augmentation
- Geometric: rotation, flip, crop, resize, affine
- Color: brightness, contrast, saturation, hue
- Advanced: MixUp, CutMix, CutOut, mosaic
- Library: albumentations (recommended), torchvision.transforms
Training Strategy
- Transfer learning: freeze backbone, train head first, then fine-tune
- Learning rate: cosine annealing, warm-up, one-cycle
- Small dataset: heavy augmentation, pretrained backbone, progressive resizing
Evaluation
- Accuracy, per-class accuracy, top-5 accuracy
- mAP (detection), IoU (segmentation)
- Confusion matrix, per-class precision/recall
- GradCAM for visual explanations
Usage Examples
"Build an image classifier for 10 product categories with only 500 images""Set up object detection for detecting defects in manufacturing images"Output Format
- Architecture: Model design with layer details
- Training Code: PyTorch / torchvision / ultralytics implementation
- Augmentation Pipeline: albumentations configuration
- Evaluation: Metrics and GradCAM visualizations
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ericwang915
- Source: ericwang915/data-scientist-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.