Nvflare convert huggingface vs Nvflare convert pytorch
Nvflare convert huggingface is an open-source agent skill for ai models and llm tooling. Nvflare convert pytorch is an open-source agent skill for ai models and llm tooling.
- Nvflare convert huggingface is best for make my hugging face trainer fine-tuning script run across several federated sites.
- Nvflare convert pytorch is best for convert my pytorch training script into a federated learning job.
- What it is
- Nvflare convert huggingface is an open-source agent skill for ai models and llm tooling.Nvflare convert pytorch is an open-source agent skill for ai models and llm tooling.
- Best for
- Make my Hugging Face Trainer fine-tuning script run across several federated sitesConvert my PyTorch training script into a federated learning job
- Kind
- SkillSkill
- Pricing
- Open sourceOpen source
- Platforms
- Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, GooseClaude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, Goose
- GitHub stars
- 4K4K
- Last commit
- Nov 2026Nov 2026
- License
- Apache-2.0Apache-2.0
- Open source
- Good for
- Make my Hugging Face Trainer fine-tuning script run across several federated sites; Convert my TRL SFTTrainer LoRA training code into a federated job; Add federated FedAvg training to an existing Seq2SeqTrainer pipelineConvert my PyTorch training script into a federated learning job; Make my existing nn.Module train across several hospitals with FedAvg; Wrap my manual training loop with NVFLARE Client API
What Nvflare convert huggingface does
Trainer-to-federated conversion
Wraps existing Trainer code with flare.patch so the model exchange is handled without rewriting the training loop.
FedAvg recipe generation
Builds a FedAvg job from the maintained recipe with parameters taken from the source configuration.
PEFT and LoRA support
Preserves adapter configuration and checks that adapter keys match between the server model and the patched Trainer.
Source evaluation preservation
Keeps per-round global-model evaluation and metric names from the original Trainer setup.
Model snapshot resolution
Uses a resolver to verify local or Hub model and dataset snapshots before validation runs.
Local validation and export
Runs compile, construction and simulation checks and exports the job with reported artifact paths.
What Nvflare convert pytorch does
Client API training conversion
Turns the training and evaluation loops into a client that exchanges model parameters and metrics with NVFLARE.
FedAvg recipe construction
Builds a federated job from the NVFLARE PyTorch FedAvg recipe, with explicit model class and arguments.
Constructor argument auditing
Checks the model's __init__ values from source so the job can reconstruct the model reliably.
Staged validation ladder
Runs compile checks, recipe construction and a local run, stopping at the first failure.
Step-weighted aggregation
Reports completed local optimizer steps per round as the FedAvg aggregation weight.
Safe checkpoint loading
Requires checkpoints to load with weights_only=True or asks for direction.
Nvflare convert huggingface vs Nvflare convert pytorch: questions
Is Nvflare convert huggingface or Nvflare convert pytorch cheaper?
Both are open source. Compare current plans on each official site.
What is the difference between Nvflare convert huggingface and Nvflare convert pytorch?
Nvflare convert huggingface is an open-source agent skill for ai models and llm tooling. Nvflare convert pytorch is an open-source agent skill for ai models and llm tooling.
