HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

ultralytics 8.2.23 New YouTube Shorts inference (#13150)

Browse source 
Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
Co-authored-by: Erfan Zekri Esfahani <40582518+eze1376@users.noreply.github.com>
This commit is contained in:
Glenn Jocher 2024-05-27 22:35:45 +02:00 • committed by GitHub
parent c371c953d5
commit 22de23ec8d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 34 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

4
docs/en/guides/triton-inference-server.md
View file

@ -62,8 +62,9 @@ The Triton Model Repository is a storage location where Triton can access and lo
from pathlib import Path
# Define paths
model_name = "yolo"
triton_repo_path = Path("tmp") / "triton_repo"
triton_model_path = triton_repo_path / "yolo"
triton_model_path = triton_repo_path / model_name
# Create directories
(triton_model_path / "1").mkdir(parents=True, exist_ok=True)
@ -86,6 +87,7 @@ The Triton Model Repository is a storage location where Triton can access and lo
Run the Triton Inference Server using Docker:
```python
import contextlib
import subprocess
import time

2
docs/en/integrations/tensorrt.md
View file

@ -125,7 +125,7 @@ The arguments provided when using [export](../modes/export.md) for an Ultralytic
- `workspace` : Controls the size (in GiB) of the device memory allocation while converting the model weights.
- Aim to use the <u>minimum</u> `workspace` value required as this prevents testing algorithms that require more `workspace` from being considered by the TensorRT builder. Setting a higher value for `workspace` may take **considerably longer** to calibrate and export.
- Adjust the `workspace` value according to your calibration needs and resource availability. While a larger `workspace` may increase calibration time, it allows TensorRT to explore a wider range of optimization tactics, potentially enhancing model performance and accuracy. Conversely, a smaller `workspace` can reduce calibration time but may limit the optimization strategies, affecting the quality of the quantized model.
- Default is `workspace=4` (GiB), this value may need to be increased if calibration crashes (exits without warning).