Search results

Is dynamic weight update / fine-tuning supported in QNN / XNNPACK backends?

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

Status: Open.

#16123 In pytorch/executorch;

· qqqqqqqwy opened on Dec 7, 2025

Support for gradients to user inputs in TrainingModule

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

Status: Open.

#13522 In pytorch/executorch;

· pkmandke opened on Aug 19, 2025

Implement load_into for the shared_ptr data loader

good first issueGood for newcomersGood for newcomers

module: runtimeIssues related to the core runtime and code under runtime/Issues related to the core runtime and code under runtime/

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

Status: Open.

#11562 In pytorch/executorch;

· JacobSzwejbka opened on Jun 11, 2025

Advice on how to run the training example in Android

module: androidIssues related to Android code, build, and executionIssues related to Android code, build, and execution

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

Status: Open.

#10593 In pytorch/executorch;

· YuanTingHsieh opened on Apr 30, 2025

update demo scripts to use .ptd

CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

fb-exported

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

release notes: training

stalePRs inactive for over 60 daysPRs inactive for over 60 days

Status: Open (in progress).

pytorch/executorchnumber 8834

#8834 In pytorch/executorch;

· JacobSzwejbka opened on Feb 28, 2025

Python apis for loading and saving .ptd from dictionaries

good first issueGood for newcomersGood for newcomers

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#8542 In pytorch/executorch;

· JacobSzwejbka opened on Feb 18, 2025

Simple ConvNet causes <code>mismatched dtypes</code> during <code>to_edge()</code> call

module: trainingIssues related to training models on edge devicesIssues related to training models on edge devices

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

Bug

#8206 In pytorch/executorch;

· holgerroth opened on Feb 5, 2025