Research · Skill
GPU Optimization for Python with NVIDIA
GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS, cuCIM, KvikIO, Warp, Newton, Numba-CUDA, or RAFT questions; and profiling, memory-transfer, kernel, or multi-GPU bottlenecks. Also use when large data-parallel Python code is slow and GPU acceleration is a plausible option, even if the user does not name CUDA.
Open in PhởWhat the skill covers
The sections of its instructions, in order.
- 01 When This Skill Applies
- 02 Choose the Smallest Suitable Layer
- 03 Optimization Workflow
- 1. Define the contract and baseline
- 2. Check suitability before porting
- 3. Try the least disruptive implementation
- 4. Keep a coherent GPU data path
- 5. Validate semantics before speed
- 6. Benchmark GPU code correctly
- 7. Keep, revise, or reject the port
- 04 Important Notes
- 05 Reference Files
- 06 Citing Scientific Agent Skills
Reference files
15 files Phở can open while it uses the skill.
- references/code_transformation_patterns.md
- references/cucim.md
- references/cudf.md
- references/cugraph.md
- references/cuml.md
- references/cupy.md
- references/cuspatial.md
- references/cuvs.md
- references/cuxfilter.md
- references/decision_framework.md
- references/installation.md
- references/kvikio.md
- references/numba.md
- references/raft.md
- references/warp.md
More in Scientific Python
- CZ CELLxGENE Census
Queries the CZ CELLxGENE Census programmatically for versioned public single-cell and spatial transcriptomics data. Use when you need population-scale cell metadata, gene…
- Cirq - Quantum Computing with Python
Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for…
- Datamol Cheminformatics Skill
Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors…
- Histolab
Extracts and preprocesses whole-slide histology image tiles with Histolab. Use for WSI inspection, tissue masks, random/grid/score-based tile extraction, H&E stain normalization…
- LaminDB
Manages biological datasets and models with LaminDB, including artifact registration, lineage tracking, schema validation, Bionty ontology annotation, query/search, collections…
- Molecular Dynamics
Runs and analyzes molecular dynamics simulations with OpenMM and MDAnalysis. Sets up protein/small molecule systems, defines force fields, runs energy minimization and production…
- Molfeat — Small-molecule featurization
Featurizes small molecules with Molfeat for QSAR/QSPR, chemical similarity, virtual screening, and molecular ML. Covers ECFP/MACCS fingerprints, RDKit descriptors…
- PennyLane
Builds and differentiates PennyLane quantum circuits, hybrid PyTorch or JAX models, molecular VQE and QAOA workflows. Use for variational quantum algorithms, quantum machine…
- polars-bio
Performs genomic interval overlap, nearest, merge, coverage, complement and subtraction on Polars DataFrames, and reads or writes BED, VCF, BCF, BAM, CRAM, GFF, GTF, FASTA and…
- TileDB-VCF
Stores and retrieves genomic variant calls with TileDB-VCF. Use for indexed single-sample VCF/BCF ingestion, incremental cohorts, region and sample queries, streaming results…
- PyTorch Geometric (PyG)
Supports PyTorch Geometric (PyG) graph neural networks — node/link/graph classification, message passing (GCN, GAT, GraphSAGE, GIN), heterogeneous graphs, neighbor sampling, and…
- TorchDrug
Builds and troubleshoots TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein…
- Transformers
Hugging Face Transformers for loading Hub models, running pipeline inference, text generation, and Trainer fine-tuning on NLP, vision, audio, and multimodal tasks. Applies when…
- Vaex
Processes large tabular scientific datasets with Vaex expressions, filtered views, streamed statistics, binned visualizations, and file conversion. Use for larger-than-RAM HDF5…
- Zarr Python
Stores and queries chunked N-D scientific arrays with Zarr-Python 3, including codecs, sharding, S3/GCS storage, and NumPy/Dask/Xarray integration. Use for array layout, bounded…
Similar research skills
Frequently asked questions
What does GPU Optimization for Python with NVIDIA do?
GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS, cuCIM, KvikIO, Warp, Newton, Numba-CUDA, or RAFT questions; and profiling, memory-transfer, kernel, or multi-GPU bottlenecks. Also use when large data-parallel Python code is slow and GPU acceleration is a plausible option, even if the user does not name CUDA.
Is GPU Optimization for Python with NVIDIA free to use?
Yes. GPU Optimization for Python with NVIDIA is published by K-Dense under the MIT licence, and you can use it with a free Phở account.
How do I use GPU Optimization for Python with NVIDIA in Phở?
Press Open in Phở, then Try to apply it to your next message, or add the Scientific Python set so Phở loads it whenever a request matches.
Does GPU Optimization for Python with NVIDIA need other tools?
No. It works in a normal Phở chat with what you type or upload.
Copied unchanged from K-Dense-AI/claude-scientific-skills at commit a25cf9e and credited to K-Dense, who is not affiliated with Phở. The full instructions and licence text ship with the skill inside Phở.