Comments (9)
Hi @HnH19
Please, could you specify what error message you are receiving?
Also, please, could you attach to your message a sample of a detection file and its associated ground-truth file?
Thanks
from review_object_detection_metrics.
Hi,
The error message if the following :
%%
No File was found
No file was found for the selected detection format in the annotations directory.
%%
Here are the annotations for a ground truth-detection pair.
By the way, there is nothing wrong with my ground truth upload, it is working.
The error occurs with the detection annotations file.
ground_truth.txt
detection.txt
Thank you in advance.
HnH
from review_object_detection_metrics.
I meet the same problem as @HnH19. The format for ground truth and predictions is same also.
from review_object_detection_metrics.
Hi @HnH19,
I tried to replicate your problem, but no error was shown. See in the images below how I did with the files you provided
Please, make sure you are following the steps below:
- Put all your groundtruths .txt files in the same folder (ex:
groundtruths/
) - As your groundtruth bounding boxes are in relative formats (Yolo (.txt)), the images are required as the width and height of the images are needed to calculate the position of the bounding boxes. Thus, put all your images in the same folder (ex:
images/
) - Create a .txt file, where each line contains the name of your classes. Example: the first line (line 0) must have the name of the objects represented by class_id=0.
- Put all your detections .txt files in the same folder. (ex:
detections\
)
Make sure that the detections, images and groundtruth files of the same image are named the same. Example: Bounding box detections of image /images/image_A.jpg
are represented by the file /detections/image_A.txt
, and their groundtruth bounding boxes are in the file /groundtruths/image_A.txt
.
Find in the file below an example I made with the files you provided. As I did not have your image, I created a dummy 100x100 black image.
HnH19.zip
Let me know if that works. :)
from review_object_detection_metrics.
@maketo97 , please, see the answer above and check if you are doing all steps correctly.
Let me know if that works now ;)
from review_object_detection_metrics.
Hi @rafaelpadilla, I was seeing the same error. However, when I use your provided data in HnH19.zip
and follow the above just as you describe the error no longer appears but the output metrics are all 0. Any thoughts?
COCO METRICS:
AP: 0.0
AP50: 0.0
AP75: 0.0
APsmall: nan
APmedium: 0.0
APlarge: nan
AR1: 0.0
AR10: 0.0
AR100: 0.0
ARsmall: nan
ARmedium: 0.0
ARlarge: nan
PASCAL METRIC (AP per class)
person: 0
PASCAL METRIC (mAP)
mAP: 0.0
from review_object_detection_metrics.
Hi @tylertroy ,
Thank you for your message. I attached the file with a small change, and that's why the results are all 0. See the explanation:
It is because the groundtruth file img_A.txt reports one object of class "person" (0)
0 0.573211 0.407236 0.347754 0.628942
and the detection img_A.txt reports a detection of object class "dog" (1)
1 0.92 0.566021 0.417603 0.544014 0.666667
Even though the IOU is high, the class of the detected object is different than the groundtruth object. So, the results must be indeed 0.
If you change the class of one of the files, you will see that the results will change :). Just change the ground-truth file img_A.txt to:
1 0.573211 0.407236 0.347754 0.628942
and it will work.
Best regards
from review_object_detection_metrics.
Closing issue as no update was reported
from review_object_detection_metrics.
Apologies @rafaelpadilla . Worked as intended.
from review_object_detection_metrics.
Related Issues (20)
- Equation (9) and (11) in publication HOT 3
- how to use commond line ? HOT 1
- Do you have the detection results of different detection models in coco json format. HOT 1
- could't open detected the annotation file made by yolov4 darknet HOT 2
- How to calculate the general Precision x recall curve of the detection model? HOT 1
- Coordinates format has repeated option HOT 4
- Can you share commands on how to run the coco and PascalVOC evaluators via terminal to generate the outputs? HOT 3
- No results HOT 4
- Feature request: Docker containerization HOT 3
- move(self, int, int): argument 1 has unexpected type 'float' HOT 5
- ModuleNotFoundError: No module named 'src' HOT 4
- Bug in the converter.py while looking for images folder ? HOT 1
- AttributeError: 'list' object has no attribute 'items' HOT 1
- Please provide command line to evaluate on terminal HOT 1
- Issue during start: TypeError: arguments did not match any overloaded call HOT 3
- np.bool is deprecated HOT 5
- numpy: set_window_title moved to manager HOT 3
- Building the conda environment causes an endless loop HOT 2
- Result is 0 HOT 3
- Getting precision, recall, F1 values 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 review_object_detection_metrics.