Comments (6)
Thank you for the help. I appreciate the work you guys are putting into this.
from ultralytics.
@initdebugs hey, right now the local weights best.pt and last.pt are updated after every epoch and artifacts are logged to supported logging platforms only after training finishes and models are stripped. Do you want a similar logging feature to log checkpoints in order to save progress?
We are working on better online logging support through HUB using which you can resume training from anywhere, even a different system. If your use case is different, please let us know so that we can support it better. thanks!
from ultralytics.
Okay that seems odd. For me it saved a best.pt and a last.pt after the first epoch. After that it hasn't updated those files. It's currently at epoch 78 but hasn't saved it.
from ultralytics.
@initdebugs okay that is not expected. It might be a bug. Can you confirm if the ckpt have not been updated by doing this.
import torch
ckpt = torch.load("last.pt")
print(ckpt["epoch"])
Thanks!
from ultralytics.
It shows the correct epoch when doing that. Does that mean that it does save it? The 'last modified' datetime for the file hasn't changed.
from ultralytics.
@initdebugs yeah it shows the epoch it was saved on. Maybe it's something with how OS's file managers check last changed? But it should not be something to worry about
from ultralytics.
Related Issues (20)
- yolo-world export onnx HOT 1
- Combine Segmentation and Detection model (Yolov7 & Yolov8) HOT 2
- training does not start HOT 1
- Using data augmentation on YOLOv8-pose HOT 3
- FPS drop while using Ultralytics HOT 2
- numpy.linalg.LinAlgError: 2-th leading minor of the array is not positive definite Error? HOT 4
- 버전 차이 질문 HOT 2
- model.engine speed is slowest than model.pt HOT 2
- Significant Drop in Performance when Switching between YOLOv8n-seg Models HOT 3
- How to crop detected object from image using YOLOv8 model without saving it. HOT 2
- validation on imgsz of 13792 HOT 3
- rtdetr weight problem HOT 5
- Limit the class of my prediction? HOT 1
- Loosing the pretrained model weights when using a new data to retrain the already trained model. HOT 7
- Training issue with latest version of ultralytics HOT 2
- export openvino with `--static_shape` when `int8=true` HOT 2
- Loss of "iscrowd" annotations when converting COCO dataset to YOLOv8 dataset HOT 6
- Multi-channel images training HOT 4
- Change trainer.py inside engine
- Example for YOLO-World - ONNX HOT 2
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.