Comments (6)
i also met issues when training:
i can't understand the reason,why its a invalid directory path?
from yolov6.
The first problem might be due to the annotation's cache file. You can delete previously generated cache files and regenerate them when using a new training set. We will also optimize this part of the settings as soon as possible, and it is expected to be completed next week. At that time, you can check whether the problem still exists.
The second question suggests that you try using absolute paths in coco.yaml to see if the same problem still exists.
Thank you for your attention
from yolov6.
i also met issues when training:
i can't understand the reason,why its a invalid directory path?
I think you can try to modify the root : from "../coco/images/train2017" to "./coco/images/train2017"
from yolov6.
I also met this issues.
This file".\runs\train\exp\predictions.json" may have problem, such as "image_id": 1333", my dataset don't have an image named 1333. How to solve it?
from yolov6.
I use the following code for training python tools/train.py --batch 4 --conf configs/yolov6s.py --data ./data/coco.yaml --device 0 --workers 0 But this mistake happened during the val AssertionError: Results do not correspond to current coco set
I am having the same issue at the moment.I have deleted the generated annotations folder and .train.json and tried again but I got the same error.Also after deleting the annotations folder and json files, even though all the paths are correct, at first I receive invalid directory path issue, then I run the command again without changing anything it trains for one epoch and I get the same 'Results do not correspond to current coco set' error. It feels like a weird loop I cant debug.Thank you for your help.
from yolov6.
from yolov6.
Related Issues (20)
- map为0 HOT 3
- distill的流程与配置 HOT 2
- Invalid label with my custom dataset from roboflow
- 自蒸馏教师模型的模型结构
- 自蒸馏损失
- Inference bottlenecked by _local_scalar_dense running on CPU
- Issue with YOLOv6 Segmentation Model Using Polygon Labels HOT 1
- YOLOv6 continues to be awesome
- How to Use QAT for Segmentation with YOLOv6? HOT 6
- Yolov6 onnx to engine file conversion
- 关于hubconf.py中process_image的疑问
- Issues with QAT Implementation for YOLO Detection Model
- How could you initialize the model to use it in code?
- 训练好的模型,infer用的best_ckpt.pt,推理出的结果没有候选框
- Empty exp, train log
- 关于自蒸馏的训练教师模型时必须开fuse ab吗
- In the classification loss, let the image ignore region not participate in the loss calculation, that is, not contribute to the classification loss
- Calculate metric failed, might check dataset. Evaluating by pycocotools doesn't work either.
- Meet with Ultralytics Team in Beijing & Shanghai - August 2024
- 如何计算模型的参数量与模型的复杂度?
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 yolov6.