efficientnet_b1 — ExecuTorch XNNPACK
efficientnet_b1_xnnpack_fp32.pte (31.2 MB, fp32, XNNPACK-delegated)
- Source: torchvision efficientnet_b1 IMAGENET1K_V2
- License: BSD-3-Clause
- Input: [[1, 3, 240, 240]] — RGB, ImageNet norm, 240x240
- Output: ImageNet logits [1,1000]
Verification (Mac arm64, executorch 1.4.0, torch 2.13.0)
Parity vs torch fp32 eager on random input:
| output | shape | max_abs_diff | corr |
|---|---|---|---|
| 0 | [1, 1000] | 6.676e-06 | 1.000000 |
Median latency over 10 runs (single Mac process, reference only — device numbers to follow): ExecuTorch 9.2 ms vs torch eager 345.4 ms.
Conversion
torch.export -> to_edge_transform_and_lower(XnnpackPartitioner) -> .pte (conversion scripts: executorch-models)
Part of executorch-models — a verified .pte zoo for ExecuTorch. Conversion scripts and all models are indexed there.
- Downloads last month
- -