Comments (2)
@alsozatch thank you for your feedback on the TensorRT integration documentation. We appreciate your insights and agree that clarity is essential for users to make informed decisions about the workspace
configuration during INT8 quantization.
We'll update the documentation to reflect the trade-offs between calibration time and model performance more clearly. Here's a revised version of the tip:
"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."
Thank you for helping us improve our documentation! 🚀
from ultralytics.
@alsozatch see #13138.
Please submit a PR directly in the future yourself if you spot room for improvement in the docs, thank you!
from ultralytics.
Related Issues (20)
- Thanks for your excellent work, We really want use yolo-world do more excellent work, can we use yolo-world trained model as base, providing embedding to fine-grained differentiation of samples for downstream tasks? HOT 1
- High CPU usage with Google Coral tflite model HOT 4
- yolov10-SCDown return error HOT 4
- Questions about using segmentation in TensorRT HOT 5
- video Inference is too slow in realtime HOT 2
- KeyError: 'Silence' while training YOLOv9 HOT 3
- Can .val use test data from custom yaml for evaluation? HOT 2
- Clarification on YOLOv8 fine-tuning HOT 12
- Get masks from model output0 and output1 HOT 13
- Confusion metrix HOT 3
- 'list' object has no attribute 'masks' HOT 6
- Dataset not found ⚠️, missing path HOT 12
- Trying to show the XY value for detecting objects on real-time
- Request for mAP of different scale HOT 15
- FATAL ERROR! reclaim_blob_allocator get wild allocator in Jetson Nano with NCNN Inference HOT 4
- data lable wrong when train yoloworld HOT 19
- Trained YOLOv8 model converted to CoreML doesn't give any predictions HOT 10
- About glean-t and yolov9-t HOT 6
- When I install torch_image, imgsz doesn't work. HOT 3
- Train subclass in Coco data set HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ultralytics.