Commit History
Upgrade onnxsim to v0.4.1 (#8632)
92e47b8
unverified
Assert `--optimize` not used with cuda device (#8569)
f4b0568
unverified
XML export `--half` fix (#8522)
c215e87
unverified
Add `--half` arguments to export.py Usage examples (#8516)
63ba0cb
unverified
Add `--hard-fail` argument to benchmarks for CI errors (#8513)
3e54651
unverified
Fix ONNX `--dynamic` export on GPU (#8378)
0c13240
unverified
Fix FP32 TensorRT model export (#8046)
6dd6aea
unverified
Remove `formats` variable to avoid `pd` conflict (#7993)
9455796
unverified
YOLOv5 Apple Metal Performance Shader (MPS) support (#7878)
c215878
unverified
OpenVINO metadata fix2 (#7954)
27911dc
unverified
OpenVINO metadata fix (#7952)
0dd66e2
unverified
Add OpenVINO metadata to export (#7947)
a3a652c
unverified
Add `--keras` argument for TF exports (#7921)
547c89b
unverified
Removed shell=True from subprocess commands that require user inputs (#7875)
fe1b503
unverified
Refactor collections and fstrings (#7821)
f000714
unverified
Rename `utils/datasets.py` > `utils/dataloaders.py` (#7799)
9d8ed37
unverified
FROM nvcr.io/nvidia/pytorch:22.04-py3 (#7680)
d4ea61e
unverified
Disallow `--dynamic` when `--half` is passed (#7669)
bff6e51
unverified
Add `--half` support for OpenVINO exports (#7615)
177da7f
unverified
Restrict TRT autoinstall to Linux-only (#7549)
1f1ec1c
unverified
Update check_requirements() with `cmds=()` argument (#7543)
f62609e
unverified
Autoinstall TensorRT if missing (#7537)
cc1d7df
unverified
Add `--half` support for FP16 CoreML exports with (#7446)
7926afc
unverified
Remove `tensorrt` pip install check (#7439)
3eefab1
unverified
Fix EdgeTPU output directory (#7399)
2da2466
unverified
Remove OpenVINO ONNX `opset<=12` check (#7398)
5333b55
unverified
Add ONNX export metadata (#7353)
3bb233a
unverified
Rename 'MacOS' to 'macOS' (#7349)
446e6f5
unverified
Update export.py (#7301)
d257c75
unverified
Context manager `open(file) as f` fixes (#7289)
5f97001
unverified
Enable TensorFlow ops for `--nms` and `--agnostic-nms` (#7281)
8d0291f
unverified
TorchScript single-output fix (#7261)
8bc839e
unverified
Export single output only (#7259)
05cf0d1
unverified
Export with official `nn.SiLU()` (#7256)
ffcbd8c
unverified
SavedModel TF Serve Fix (#7228)
734ab03
unverified
CLI `fire` prep updates (#7229)
2c32218
unverified
precommit: yapf (#5494)
c3d5ac1
unverified
Sidestep `os.path.relpath()` Windows bug (#7158)
e19f87e
unverified
Consistent saved_model output format (#7032)
4effd06
unverified
Mrinal Jain
commited on
YOLOv5 Export Benchmarks for GPU (#6963)
932dc78
unverified
pt model to cpu on TF export
99de551
unverified
EdgeTPU optimizations (#6808)
c13d4ce
unverified
Fix TRT `max_workspace_size` deprecation notice (#6856)
acc58c1
unverified
Default FP16 TensorRT export (#6798)
596de6d
unverified
Fix export for 1-channel images (#6780)
b2adc7c
unverified
Louis Combaldieu
commited on
