“cuda”的搜索结果

Train and evaluate robot policies, the models that choose actions, using RSL-RL or RLinf in healthcare simulations. Export trained artifacts and check that they work through the normal Workflow runner.

中文简介正在准备中。当前显示英语简介。

• Train an ultrasound probe reach policy
• Post-train a Trocar policy from a local GR00T checkpoint
• Evaluate and export an existing training checkpoint

Write, review, and migrate DALI data processing code using dynamic mode, which calls operators directly from ordinary Python. Handle readers, batches, GPU processing, PyTorch conversion, random state, and checkpoints with the conventions specific to this API.

中文简介正在准备中。当前显示英语简介。

• Writing image preprocessing in ordinary Python loops
• Migrating a DALI pipeline to dynamic mode
• Reviewing batch slicing and PyTorch conversion

Move pandas-style table processing to NVIDIA GPUs using cuDF. Choose minimal-change acceleration or explicit GPU code, handle joins and aggregations, compare results with pandas, and use dask-cuDF when data exceeds GPU memory.

中文简介正在准备中。当前显示英语简介。

• Accelerate existing pandas processing with minimal changes
• Move joins and aggregations to GPU code
• Compare missing values and ordering with pandas

Build video analytics applications with NVIDIA DeepStream and Python's pyservicemaker API. The skill guides video input, object detection, optional tracking, annotated output, and Kafka messaging, with configuration guidance and fixes for common pipeline errors.

中文简介正在准备中。当前显示英语简介。

• Detect objects in video files or camera streams
• Track objects across video frames
• Send detection results to Kafka

Continue training a KERMT molecular model on a custom SMILES corpus, where text strings describe molecular structures. The skill validates data and checkpoints, prepares training inputs, and launches containerized training with saved logs and checkpoints.

中文简介正在准备中。当前显示英语简介。

• Continue molecular model pretraining on a custom SMILES corpus
• Start additional training with a new learning-rate schedule
• Resume an interrupted training run from a saved checkpoint

Generate molecular predictions from a CSV of SMILES strings, a text notation for molecular structures, using a finetuned KERMT model. Validate inputs, clean data, compute molecular features, and return predictions with execution logs and a replay command.

中文简介正在准备中。当前显示英语简介。

• Predict targets for molecules listed in a SMILES CSV
• Check whether a saved model is suitable for inference
• Reuse prepared data for another prediction run

Fine-tune Parakeet TDT v2 on clinical vocabulary using NeMo, then compare key-term error rates before and after training. The results guide whether to keep the model, expand the dataset, or prepare it for deployment.

中文简介正在准备中。当前显示英语简介。

• Reducing transcription errors in medication names
• Checking whether training improved clinical term recognition
• Deciding whether to expand the training dataset

Extend an existing grover_base model with a new decoder that reconstructs molecular structure strings, then continue pretraining on your corpus. The workflow validates inputs, rebuilds vocabulary components, and records the conversion and training outputs.

中文简介正在准备中。当前显示英语简介。

• Extend an existing grover_base model with cMIM
• Continue pretraining on a molecular corpus
• Trace checkpoint conversion and training records

Fine-tune a pretrained KERMT model on a CSV of molecular structures and labeled targets. The workflow checks inputs, prepares features, launches training in a container, and saves trained models, logs, and test results.

中文简介正在准备中。当前显示英语简介。

• Train a molecular model on labeled CSV data
• Select specific prediction targets for training
• Reuse prepared data to compare training settings

Pretrain a new KERMT chemistry model from a molecular corpus, starting with random weights. The skill validates the data, builds fresh vocabularies, estimates runtime, and launches containerized training with logs and records for reproducing the run.

中文简介正在准备中。当前显示英语简介。

• Train a new model on a specialized chemistry corpus
• Run pretraining without an existing checkpoint
• Choose a training objective and batch size

Start a cuOpt REST server and build Python or curl clients to submit routing and optimization problems. Follow result polling, translate Python API field names into REST fields, and diagnose rejected requests.

中文简介正在准备中。当前显示英语简介。

• Start a cuOpt server locally or in Docker
• Submit a routing problem from Python
• Retrieve optimization results by polling

Turn molecules listed in a SMILES CSV into numerical feature vectors using a KERMT model. The workflow checks the model and input, cleans molecular notation, and saves four sets of vectors with validity information.

中文简介正在准备中。当前显示英语简介。

• Extract molecular feature vectors from a SMILES CSV
• Reuse a local KERMT checkpoint
• Extract features with a released pretrained model

Generate, verify, refine, and accelerate C/C++ or CUDA code from MATLAB with MATLAB Coder, Embedded Coder, GPU Coder, or MATLAB Test. Also covers writing codegen-ready MATLAB code: language constraints, coder.* directives, and optimization patterns. Triggers on: codegen, MEX, deploy MATLAB as C/C++, GPU Coder, coder.screener, coder.config, coder.gpuConfig, coder.typeof, coder.runTest, matlabtest.coder.TestCase, SIL, embedded config, no dynamic memory, EnableMexProfiling, coder.timeit, coder.perfCompare, %#codegen, writing codegen-ready MATLAB, code generation readiness, coder.varsize, coder.unroll, coder.noImplicitExpansionInFunction, coder.ceval, coder.inline, coder.extrinsic, coder.const, coder.classSignature, class codegen limitations, temporal types codegen, DMA-off, stack-only, host-target InstructionSetExtensions, SIMDAcceleration, OptimizeReductions, host SIMD tuning, host OpenMP, codegen performance.

中文简介正在准备中。当前显示原文说明。

Deploy AI models to embedded hardware using MathWorks tools (MATLAB, Simulink, Embedded Coder). Covers two workflow patterns: (1) MathWorks-native or imported models rebuilt as dlnetwork for lean hardware, (2) direct C/C++ code generation from PyTorch and LiteRT models. Both patterns support all targets (Cortex-M/A/R, x86, GPU). Trigger when: user wants to deploy AI to embedded targets; generate C/CUDA from neural networks; compress AI models for MCU; integrate AI in Simulink for system-level simulation; import PyTorch/ONNX/TensorFlow models for embedded deployment; optimize AI for resource-constrained hardware; or use loadPyTorchExportedProgram, loadLiteRTModel, importNetworkFromPyTorch, importNetworkFromONNX, importNetworkFromTensorFlow, importNetworkFromKeras, dlquantizer, exportNetworkToSimulink, or Embedded Coder with AI models.

中文简介正在准备中。当前显示原文说明。

Set up and manage NVIDIA GPU servers for AI workloads — driver installation, CUDA toolkit, container toolkit, MIG partitioning, GPU health monitoring, and multi-GPU configuration for LLM inference and training.

中文简介正在准备中。当前显示原文说明。

Diagnose NCCL failures and adjacent training-pod failures on HyperPod GPU clusters (EKS or Slurm) — training hangs, AllReduce / collective-op timeouts, EFA or libfabric errors, rendezvous failures, EFA TCP fallback, /dev/shm or memlock issues, NCCL version mismatch across pods, container OOM / exit-137 / OOMKilled, GPU OOM (CUDA out of memory), CrashLoopBackOff / Pending pods, MASTER_ADDR DNS, NetworkPolicy blocking. Not for single-node hardware faults (→ hyperpod-node-debugger § G) or cluster-creation EFA / SSM failures (→ hyperpod-cluster-debugger § A / § F).

中文简介正在准备中。当前显示原文说明。

Install Triton + SageAttention to accelerate ComfyUI (the sageattn attention_mode and inductor torch.compile used by WanVideoWrapper / many video graphs). Windows-first (triton-windows + woct0rdho prebuilt SageAttention wheels matched to torch/CUDA/python into the RIGHT python), plus Linux (official triton + build) and Mac (N/A → sdpa/MPS). Also covers the SAFE sdpa / no-compile fallback so an example that assumes sageattn + torch.compile still runs when these aren't installed (video-extend TRAP 5). Use when a loader crashes with "No module named 'sageattention'" or reports triton unavailable, when asked to speed up Wan/video workflows, or when deciding whether to install acceleration vs. fall back.

中文简介正在准备中。当前显示原文说明。

Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving.

中文简介正在准备中。当前显示原文说明。

Implement or debug Hugging Face Spaces ZeroGPU code using `@spaces.GPU`, including isolation, state, duration, sizing, and CUDA-wheel constraints. Use for ZeroGPU-specific behavior; use huggingface-spaces for general hosting.

中文简介正在准备中。当前显示原文说明。

Expert guide for Backend.AI distributed computing platform. Automatically activates when users ask about: - Backend.AI architecture, components (Manager, Agent, Storage Proxy, Webserver, App Proxy) - Features (session scheduling, Sokovan orchestrator, multi-tenancy, resource allocation) - APIs (REST, GraphQL), authentication, RBAC authorization - Container runtime (kernels, jail sandbox, hook library, virtual folders) - Accelerator support (CUDA, ROCm, TPU, NPU, Graphcore IPU) - Client SDKs (Python, Java, JavaScript, PHP) - Setup, requirements (Python 3.13+, Docker, PostgreSQL, Redis, etcd) - How WebUI connects to/interacts with Backend.AI backend - Plugin interfaces, development setup, infrastructure Use when user mentions "Backend.AI", "backend.ai", "Sokovan", component names, or asks about the backend platform this WebUI connects to. allowed-tools: WebFetch, Read

中文简介正在准备中。当前显示原文说明。

Convert PyTorch AT_DISPATCH macros to AT_DISPATCH_V2 format in ATen C++ code. Use when porting AT_DISPATCH_ALL_TYPES_AND*, AT_DISPATCH_FLOATING_TYPES*, or other dispatch macros to the new v2 API. For ATen kernel files, CUDA kernels, and native operator implementations.

中文简介正在准备中。当前显示原文说明。

Routes official FlashVSR diffusion-based streaming video super-resolution setup, model preparation, and CUDA inference for v1 and v1.1.

中文简介正在准备中。当前显示原文说明。

Use Gaussian-SLAM for CUDA RGB-D dense SLAM, Gaussian-splatting mapping, dataset/configuration preflight, checkpoint evaluation, reconstruction metrics, and global-map refinement.

中文简介正在准备中。当前显示原文说明。

Use the faster-whisper package for CTranslate2-backed Whisper transcription, model selection, CPU/CUDA setup, audio utilities, VAD, timestamps, and conversion guidance.

中文简介正在准备中。当前显示原文说明。