quickstart/

[giscus[bot]]

quickstart/

Explore various methods to install Ultralytics using pip, conda, git and Docker. Learn how to use Ultralytics with command line interface or within your Python projects.

https://docs.ultralytics.com/quickstart/

[Debawy]

Can you add this commands to relevant files

import os
os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"

to work around this error

OMP: Error #15: Initializing libiomp5md.dll, but found libiomp5md.dll already initialized.
OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see http://www.intel.com/software/products/support/.

To Reproduce

[glenn-jocher]

Thanks for pointing that out! While setting the KMP_DUPLICATE_LIB_OK environment variable can indeed provide a workaround for the multiple copies of the OpenMP runtime error, it's generally recommended to address the root cause by ensuring only a single OpenMP runtime is linked, as indicated by the error message itself. 🛠️

Injecting this workaround directly into the Ultralytics files might inadvertently mask underlying issues for some users. For those facing this specific error and understanding the associated risks, applying the workaround in their environment or at the beginning of their scripts before importing any other libraries can offer a temporary solution:

import os
os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"

Before applying this workaround, I'd suggest investigating the installation and dependencies of the libraries involved to identify possible conflicts with the OpenMP runtime. Feel free to reach out if you run into any hurdles while exploring these avenues. Happy coding! 🚀

[choiyhking]

Hello, I'm using yolov8 on NVIDIA Jetson Nano.

I executed objecdetection prediction on native, docker(Dockerfile-jetson), docker(Dockerfile-arm64).
If I use CPU, the results(inference time) are native > Dockerfile-jetson > Dockerfile-arm64.
In my opinion....It should be similar or Dockerfile-jetson faster than Dockerfile-arm64
(because jetson version is more optimized for jetson platform).
I read about Dockerfile but I can't find major difference btw Dockerfile-jetson and Dockerfile-arm64.

Can you explain about this?

Thanks.

[pderrenger]

Hey there! 👋 Thanks for reaching out with your observation regarding the inference times across different Docker environments on the NVIDIA Jetson Nano. It's an interesting point you've raised.

The Dockerfile-jetson is indeed optimized for the NVIDIA Jetson platform, potentially including optimizations specific to its GPU architecture that might not be present in the Dockerfile-arm64. However, the variance in inference times you're observing can depend on several factors:

• The base images used for each Dockerfile
• Any additional libraries or drivers included
• The specific optimization flags set during the build process

While they are both aimed at ARM64 architectures, the Dockerfile-jetson might include some Jetson-specific tweaks that aren't in the Dockerfile-arm64, which is more generic.

As a quick suggestion, you could look into ensuring both Docker environments are utilizing the GPU correctly and to its fullest potential. Sometimes, Docker configurations might not fully leverage the hardware acceleration available.

If you're continuing to see inconsistencies and would like further optimization on the Jetson platform, diving deeper into the Dockerfiles and possibly customizing to suit your exact need could be beneficial.

Hope this answers your question! Let us know if there's anything else we can do. Happy coding! 🚀

[yjh1314yjh]

1. Hello, why does the model I trained with yolov8 predict a single picture in the same scene without false detection, but it will be false detection when combined with the camera? What is the reason? 2. 800 pictures include training set, test set and verification set, but the final loss will no longer decrease when it drops to 0.7. The accuracy rate of AP50 is 95%, while the accuracy rate of AP50-95 is only 67%.

[glenn-jocher]

Hello 👋! Let's tackle your queries one at a time:

1. For the issue of false detections with a camera vs. single pictures, this could be due to differences in image quality, lighting conditions, or even the camera's real-time feed having variations that aren't present in the static images. You might want to adjust the conf (confidence threshold) parameter a bit higher when using the camera feed to reduce false positives.

2. Regarding the training plateau at a loss of 0.7, this is relatively common in deep learning. The model might have reached its capacity given the current dataset size and complexity. Since you have an AP50 of 95% but a lower AP50-95, consider augmenting your dataset or employing techniques like mixup or mosaic augmentation to introduce more variability into your training process. This could help the model generalize better and potentially improve on finer metrics like AP50-95.

For augmenting your dataset, you can explore the augmentation options provided in the Ultralytics documentation or directly within your training configuration file.

Best of luck with your model training and optimization! 🚀

[yjh1314yjh]

So, as my dataset expands from 800 images to 2000 images, is my pre-trained model better off with yolov8n or yolov8s?

[glenn-jocher]

Hey there! 😊 It looks like you're facing a choice between YOLOv8n and YOLOv8s for your expanded dataset.

1. Dataset Size: Both models can benefit from more data. However, YOLOv8n, being a lighter model, is quicker to train but may saturate in performance with more data sooner than YOLOv8s.

2. YOLOv8s: On the other hand, YOLOv8s, being slightly larger, has a higher capacity and might leverage the additional data more effectively to learn better representations, especially in a more complex or varied dataset situation.

Given this, if you’re aiming to maximize your model's performance with the larger dataset, YOLOv8s may be a better choice as it can potentially exploit the increased data diversity and volume more robustly. However, you might also want to consider factors such as inference speed and available computational resources since YOLOv8s will likely be slower than YOLOv8n in inference.

# Example command to train with YOLOv8s
yolo detect train data=your_dataset.yaml model=yolov8s.pt epochs=100

Always happy to help! Let me know if there's anything else. 🚀

[eeasel]

hi, im using yolov8 and i have some problems...
I used anaconda on pycharm for this,and i have installed ultralytics and other packages,but when i run this command:yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg',it told me:
Traceback (most recent call last):
File "\?\D:\ProgramFiles\anaconda\envs\yolov8\Scripts\yolo-script.py", line 33, in
sys.exit(load_entry_point('ultralytics==8.0.228', 'console_scripts', 'yolo')())
File "\?\D:\ProgramFiles\anaconda\envs\yolov8\Scripts\yolo-script.py", line 25, in importlib_load_entry_point
return next(matches).load()
File "D:\ProgramFiles\anaconda\envs\yolov8\lib\importlib\metadata_init_.py", line 171, in load
module = import_module(match.group('module'))
File "D:\ProgramFiles\anaconda\envs\yolov8\lib\importlib_init_.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1050, in _gcd_import
File "", line 1027, in _find_and_load
File "", line 992, in _find_and_load_unlocked
File "", line 241, in _call_with_frames_removed
File "", line 1050, in _gcd_import
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in load_unlocked
File "", line 883, in exec_module
File "", line 241, in call_with_frames_removed
File "D:\ProgramFiles\anaconda\envs\yolov8\lib\site-packages\ultralytics_init.py", line 5, in
from ultralytics.data.explorer.explorer import Explorer
File "D:\ProgramFiles\anaconda\envs\yolov8\lib\site-packages\ultralytics\data_init.py", line 3, in
from .base import BaseDataset
File "D:\ProgramFiles\anaconda\envs\yolov8\lib\site-packages\ultralytics\data\base.py", line 17, in
from ultralytics.utils import DEFAULT_CFG, LOCAL_RANK, LOGGER, NUM_THREADS, TQDM
^
i have no idea about this ┭┮﹏┭┮, could you please give me some directions?
Ummm...last week it did work,but this week when i installed pyqt5 and some tools and i try to do pyqt5 designer,it didn't work anymore,if it because of this?

[glenn-jocher]

Hey there! 👋 It seems the error you're encountering is related to a YAML scanning issue, implying there's a problem with the formatting in one of the YAML files the Ultralytics package is trying to read. These issues usually arise from syntax errors in YAML files, such as improper indentation or unescaped characters.

Without knowing the exact YAML file Ultralytics is trying to parse, it's a bit challenging to pinpoint the issue directly. However, it's not uncommon for custom configuration files (like your dataset, model config, etc.) to be the source of such errors. Here are a couple of steps you might consider:

1. Review YAML files: If you've created or modified any YAML files (e.g., for custom datasets or models), please double-check their syntax. Look closely at line 17-18, as mentioned in the error, for any syntax issues.

2. Environment Check: Although it sounds more like a YAML syntax issue, ensuring your environment is clean doesn't hurt. Creating a fresh environment and reinstalling the necessary packages could sometimes resolve unexpected issues.

3. Try Default Files: If you're working with custom YAML files, try running the command with the default files provided by Ultralytics, if you haven't already. This can help determine if the issue lies within the custom files.

# For example, using an official model and default dataset
yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'

If the problem persists even with the default configurations or you're unsure which YAML file is causing the issue, you might want to share more details about your setup or any custom YAML files you're using. That would help in diagnosing the problem more accurately! 😊

Hope this helps! Let me know if there's anything else I can assist you with.

[eeasel]

Actually,I havenot use my dataset or model,I use "from ultralytics import yolo" for this:
model = YOLO('yolov8n.pt')
results = model('https://ultralytics.com/images/bus.jpg')
but it doesn't work.I don't know which yaml file it read.

[glenn-jocher]

@eeasel hey there! 👋 It looks like you're facing an issue with YAML file parsing while trying to use YOLOv8 directly from the Ultralytics installation. This error usually occurs due to a syntax issue within a YAML file that the library tries to read upon import or execution.

Given that you mentioned not using a custom dataset or model and simply running a basic prediction, the problem might stem from a default configuration or model file. Here are a few steps you could take to troubleshoot:

• Check Your Installation: Ensure you're using the latest version of the ultralytics package. Sometimes, such issues are resolved in newer releases. You can update using pip install -U ultralytics.

• Inspect Default YAML Files: If you've interacted with any .yaml files within the library's directories or your working directory (even inadvertently), please double-check their syntax. Look out for missing colons, incorrect indentations, or misplaced elements.

• Fresh Environment Test: If possible, try creating a new Python environment and reinstall Ultralytics to see if the issue persists. This helps to rule out environment-specific problems.

conda create -n new_env python=3.8
conda activate new_env
pip install ultralytics

• Report the Issue: If the problem continues, it might be helpful for the development team to know. You could open an issue on the Ultralytics GitHub with detailed information about your setup, the exact error message, and steps to reproduce the error. They are super responsive and helpful!

It's quite puzzling that this issue sprung up without apparent changes to datasets or model configurations on your end. Rest assured, the Ultralytics team continually works on enhancing the library, and they greatly appreciate feedback from users like you to pinpoint and remedy such issues.

Let me know how these steps go, or if you've already tried some of them! 😊

[eeasel]

Thank u for continuing to reply me.I really appreciate it a lot.
I printed the yaml file's location and I found is in C:\Users\lenovo\AppData\Roaming\Ultralytics and it is settings.yaml.(I have no idea why it is located in C disk)Here are the content:
settings_version: 0.0.4
datasets_dir: D:\myproject\ultralytics-main\datasets
weights_dir: weights
runs_dir: runs
uuid: (mosaic)
sync: true
api_key: ''
clearml: true
comet: true
dvc: true
hub: true
mlflow: true
neptune: true
raytune: true
tensorboard: true
wandb: true
e

I have no idea where the "e" come from!!!But finally I have solve this problem(by delete the additional "e").
Thanks again for helping me so patient!!!

[glenn-jocher]

@eeasel hey there! I'm really glad to hear you figured out the issue with the extra "e" in the settings.yaml file 🎉. It's quite intriguing how such small things can cause big headaches, isn't it? 😅

The fact that the file was located in C:\Users\lenovo\AppData\Roaming\Ultralytics is pretty standard for configuration files, especially on Windows. It's a common practice for applications to store settings in the AppData folder to keep user-specific configurations separate from the main application files.

Deleting the problematic "e" was the right move. YAML files are quite sensitive to syntax, so even a minor typo can lead to errors. It's awesome that you managed to troubleshoot this!

If you encounter any more puzzles or questions down the road, feel free to ask. The Ultralytics community is always here to help! And thanks again for your patience and perseverance. 🚀

Happy coding with YOLOv8! 🎈

[lauramcdaniel006]

How can I implement the detection and segmentation model? I find that I can implement the detection model but not the segmentation model

[pderrenger]

Hey there! 😊 To implement a segmentation model alongside your detection model, you'll first need to ensure you have a segmentation model, like yolov8n-seg.pt, ready. Here's a quick example of how to run a segmentation model prediction:

from ultralytics import YOLO

# Load a segmentation model
model = YOLO('yolov8n-seg.pt')  # Ensure you have this segmentation model

# Perform segmentation on an image
results = model('path/to/your/image.jpg')

# You can then visualize or save the segmented image
results.show()
results.save()

If you're diving into more specific customizations or encountering any issues, feel free to share more details! 🛠️

[jonaChong70]

Hi, I am using YOLOv8 and I am facing problem.
I am using YOLO in Python. The Python program (object_detection2.py) will be invoked by c++ using Python/C API method.

Error message:
ERROR: Module not imported
Traceback (most recent call last):
File "/home/jonathanchong/object_detection2.py", line 2, in
from ultralytics import YOLO
ModuleNotFoundError: No module named 'ultralytics'

There is no error whenever I commented out YOLO command.
What should I do?

/object_detection2.py******/
from ultralytics import YOLO
#import cv2
import json

def init():
model_path = '../resources/multibox-box-detection-model.pt'
#model = YOLO(model_path)
print('Python function init() called')

def detect():
try:
init()
print('Python function detect() called')

/c++ function******/
CPyInstance pyInstance;
CPyObject pName = PyUnicode_FromString("object_detection2");
CPyObject pModule = PyImport_Import(pName);
if(pModule)
{
CPyObject pFunc = PyObject_GetAttrString(pModule, "detect");
if(pFunc && PyCallable_Check(pFunc))
{
CPyObject pValue = PyObject_CallObject(pFunc, NULL);
}
else
{
std::cout << "ERROR: function detect()\n" << std::endl;
}
}
else
{
std::cout << "ERROR: Module not imported\n" << std::endl;
}

[glenn-jocher]

@jonaChong70 hi there,

Great to hear that switching to Python 3.10 resolved the issue! It's always satisfying to pinpoint the root cause and find a solution. If you have any more questions or need further assistance as you continue working with YOLOv8, feel free to reach out. Happy coding! 😊

[jonaChong70]

Now I have this error:
[enforce fail at inline_container.cc:157] . archive does not contain any files

Whenever after these codes. Any idea?
model_path = '../resources/multibox-box-detection-model.pt'
model = YOLO(model_path)

[glenn-jocher]

@jonaChong70 hello,

It seems like the error you're encountering, [enforce fail at inline_container.cc:157] . archive does not contain any files, typically occurs when the model file you're trying to load is either corrupted or was not saved properly.

Here are a few steps you can take to resolve this issue:

1. Verify the Model File: Ensure that the model file at '../resources/multibox-box-detection-model.pt' exists and is not corrupted. You might want to try re-downloading or retraining the model if possible.

2. Check File Path: Ensure that the path to the model file is correct relative to the script that is being executed. Sometimes, path issues can cause the file not to be found, leading to errors when attempting to load it.

3. Permissions: Check if your application has the necessary permissions to read the model file.

If these steps do not resolve the issue, consider providing more details or the steps leading up to saving the model file, which might give more insights into what might have gone wrong.

[jonaChong70]

Hi Glenn.
I have verified that the file and path are fine.

1. Rename the .pt file to something else it says "No such file or directory..."
2. Same error "[enforce fail at inline_container.cc:157] . archive does not contain any files" after downloaded another copy from https://huggingface.co/Ultralytics/YOLOv8/blob/main/yolov8n.pt

I have tested by executing the "object_detection2.py" through terminal, there is no error.
So why executing .py through Python/C API method does not work?

[glenn-jocher]

Hello,

It sounds like you've done a thorough check on the file and path, which is great. Since the script runs without issues directly through the terminal but not when invoked via the Python/C API, this suggests there might be an environment discrepancy when the script is called from C++.

Here are a few things you could check or try:

1. Environment Consistency: Ensure that the Python environment being used when calling from C++ is the same as when running directly from the terminal. You might be using a different Python interpreter without realizing it.

2. Explicit Interpreter: Specify the Python interpreter explicitly in your C++ code to ensure it's using the correct one. You can do this by setting the PYTHONHOME and PYTHONPATH environment variables programmatically in your C++ code before initializing the Python interpreter.

3. Dependencies: Double-check that all necessary dependencies are installed in the Python environment used by your C++ application. Sometimes, libraries might be installed in a different environment that is not accessible by the Python instance called from C++.

4. Error Logging: Enhance the logging in your Python script to catch and log exceptions or errors that might give more insight into what fails when the script is invoked from C++.

5. Test with Minimal Code: Try running a minimal Python script that just loads the model without any other operations, to isolate whether the issue is with loading the model or some other part of the script.

If these suggestions do not resolve the issue, it might be helpful to look into how the Python interpreter is initialized in your C++ code and ensure it's configured correctly to use the correct environment and dependencies.

Let me know how it goes or if you need further assistance!

[a-akram]

Hi all,

Has anybody tried building a conda environment? I tried it by installing the latest PyTorch (CPU) for conda and then by installing Ultralytics for conda. I got a lot of conflict errors. However, installing ultralytics from PyPI inside the conda environment works fine. I think there should be recommended versions for both PyTorch and Ultralytics.

[pderrenger]

@a-akram hi there,

Thank you for reaching out! It sounds like you're encountering dependency conflicts when trying to set up a conda environment for Ultralytics. This can sometimes happen due to the complex nature of package dependencies.

To help resolve this, I recommend installing all necessary packages in a single command to allow conda to manage dependencies more effectively. Here's a command that should help:

conda install -c pytorch -c nvidia -c conda-forge pytorch torchvision pytorch-cuda=11.8 ultralytics

This command ensures that pytorch, torchvision, pytorch-cuda, and ultralytics are installed together, which can help avoid conflicts.

If you continue to experience issues, please ensure you are using the latest versions of all packages. You can check for updates and install the latest versions using:

conda update --all

For more detailed guidance, you can refer to our Quickstart Guide.

If you have any further questions or need additional assistance, feel free to ask. Happy coding! 😊

[jonaChong70]

Please be specific on the conflict errors. It will be helpful to post the errors. On Thursday 11 July 2024 at 05:42:24 pm GMT+8, Adeel Akram ***@***.***> wrote: Hi all, Does anybody tried building a conda envirionment? I tried it by first installing latest PyTorch (CPU) for conda and then by installing Ultralytics for conda. I got a lot of conflict errors. However, installing ultralytics from PyPI inside conda environemnt works fine. I think there should be recommended versions for both PyTorch and Ultralytics. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[pderrenger]

@jonaChong70 hi Adeel,

Thank you for reaching out! 😊

To address your issue with building a conda environment, it would be very helpful if you could provide specific conflict error messages you encountered. This will allow us to better understand and troubleshoot the problem.

In the meantime, here are some steps you can follow to ensure a smooth installation:

1. Install all packages together: When working in a CUDA environment, it's best practice to install ultralytics, pytorch, and pytorch-cuda in a single command. This helps the conda package manager resolve any conflicts. For example:

   conda install -c pytorch -c nvidia -c conda-forge pytorch torchvision pytorch-cuda=11.8 ultralytics

2. Verify package versions: Ensure you are using the latest versions of all packages. Sometimes, conflicts arise from outdated dependencies.

3. Reproducible example: If the issue persists, please provide a minimum reproducible example. This will help us diagnose the problem more effectively. You can find guidance on creating a reproducible example here.

If you prefer using pip within a conda environment and it works fine for you, that's also a valid approach. However, the conda installation should ideally work without issues, and we're here to help resolve any conflicts you encounter.

Feel free to share the error logs, and we'll assist you further!

[Miseson2024]

Thank you for this great application! It has been amazing detecting objects with high occlusion successfully although I am a real Python beginner.

I successfully trained a model myself using Spyder in Anaconda navigator after setting everything myself. I also successfully predicted objects in my own dataset until my Anaconda navigator stopped working when I tried to install an application at certain point. So, now I have uninstalled everything and reinstalled Python this time without Anaconda: Python and Spyder connected with kernels. Before, I did not like that I could not use pip due to Anaconda. I installed Ultralytics and PyTorch following the instruction. However, the prediction is completely wrong although the console does not show any error message. I have no clue what I should troubleshoot. What do you think is wrong?

[pderrenger]

@Miseson2024 thank you for your kind words! It sounds like you've done a great job setting up and using Ultralytics YOLO so far. Given that your predictions are now incorrect after reinstalling Python and Spyder, I recommend checking a few things:

1. Ensure you have the latest versions of Ultralytics and PyTorch installed. You can update them using pip install -U ultralytics torch.
2. Verify that your model and dataset paths are correctly specified and accessible.
3. Check if your model weights were correctly saved and loaded. Sometimes, re-training the model can help if the weights are corrupted.

If the issue persists, please try running your script outside of Spyder to see if the problem is specific to the IDE. For more detailed troubleshooting, refer to our Quickstart Guide.

[Miseson2024]

Thank you, @pderrenger!

[omrir]

I tried the basic install using 'pip install ultralytics' on computer with only CPU.
I run the test CLI --> yolo detect predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'
Got strange result. A lot of bounding boxes uncorrected on the upper part of the image:
I cannot paste an image here, so here this is the textual result:
Found https://ultralytics.com/images/bus.jpg locally at bus.jpg
image 1/1 C:\Users\Omri-Einat\Desktop\pythonProject\bus.jpg: 640x480 13 persons, 65 bicycles, 14 cars, 4 motorcycles, 4 buss, 1 train, 1 truck, 2 boats, 1 traffic light, 25 fire hydrants, 4 stop signs, 43 parking meters, 7 benchs, 3
cats, 1 horse, 1 elephant, 6 giraffes, 15 backpacks, 8 umbrellas, 53 handbags, 1 tie, 1 frisbee, 3 skiss, 2 baseball bats, 7 baseball gloves, 1 tennis racket, 1 hot dog, 1 pizza, 1 laptop, 1 remote, 4 cell phones, 1 microwave, 4 ov
ens, 1 sink, 1328.2ms
Speed: 0.0ms preprocess, 1328.2ms inference, 31.3ms postprocess per image at shape (1, 3, 640, 480)

Any idea?

Thanks!

[pderrenger]

@omrir thank you for reporting this issue. It seems like the model is producing an unusually high number of detections. This could be due to a bug or an issue with the specific model version. Please ensure you are using the latest version of the ultralytics package by running pip install -U ultralytics. If the problem persists, try using a different pre-trained model or adjusting the confidence threshold with the conf argument to filter out low-confidence detections. If the issue continues, please provide more details so we can investigate further.

[omrir]

I performed what you suggested. Also tried different model and higher confidence. Still the same result:

Successfully installed ultralytics-8.2.71

[notice] A new release of pip is available: 23.2.1 -> 24.2
[notice] To update, run: python.exe -m pip install --upgrade pip
PS C:\Users\Desktop\pythonProject> yolo detect predict model=yolov8s.pt source='https://ultralytics.com/images/bus.jpg'
Ultralytics YOLOv8.2.71 🚀 Python-3.10.7 torch-2.4.0+cpu CPU (Intel Core(TM) i5-3570 3.40GHz)
YOLOv8s summary (fused): 168 layers, 11,156,544 parameters, 0 gradients, 28.6 GFLOPs

Found https://ultralytics.com/images/bus.jpg locally at bus.jpg
image 1/1 C:\Users\Desktop\pythonProject\bus.jpg: 640x480 15 persons, 35 bicycles, 1 car, 7 motorcycles, 9 boats, 4 traffic lights, 29 fire hydrants, 2 stop signs, 18 parking meters, 3 benchs, 1 cow, 19 backpacks, 38 hand
bags, 9 frisbees, 8 skiss, 2 baseball bats, 4 baseball gloves, 3 tennis rackets, 6 forks, 1 knife, 3 spoons, 1 hot dog, 1 dining table, 75 cell phones, 1 oven, 1 clock, 2 hair driers, 2 toothbrushs, 1328.1ms
Speed: 0.0ms preprocess, 1328.1ms inference, 31.2ms postprocess per image at shape (1, 3, 640, 480)
Results saved to runs\detect\predict8
💡 Learn more at https://docs.ultralytics.com/modes/predict
PS C:\Users\Desktop\pythonProject> yolo detect predict model=yolov8m.pt conf=0.7 source='https://ultralytics.com/images/bus.jpg'
Ultralytics YOLOv8.2.71 🚀 Python-3.10.7 torch-2.4.0+cpu CPU (Intel Core(TM) i5-3570 3.40GHz)
YOLOv8m summary (fused): 218 layers, 25,886,080 parameters, 0 gradients, 78.9 GFLOPs

Found https://ultralytics.com/images/bus.jpg locally at bus.jpg
image 1/1 C:\Users\Desktop\pythonProject\bus.jpg: 640x480 9 persons, 2 bicycles, 30 cars, 20 airplanes, 13 buss, 2 boats, 7 fire hydrants, 2 parking meters, 2 benchs, 1 dog, 1 horse, 4 sheeps, 1 backpack, 57 umbrellas, 7
handbags, 1 tie, 21 frisbees, 1 skis, 1 snowboard, 2 sports balls, 4 kites, 46 baseball bats, 1 skateboard, 3 surfboards, 3 tennis rackets, 12 bottles, 2 cups, 12 forks, 4 knifes, 19 spoons, 1 banana, 2 pizzas, 3 dining tables, 1 re
mote, 2 vases, 1 toothbrush, 2906.4ms
Speed: 15.6ms preprocess, 2906.4ms inference, 0.0ms postprocess per image at shape (1, 3, 640, 480)
Results saved to runs\detect\predict9
💡 Learn more at https://docs.ultralytics.com/modes/predict

[omrir]

I also see that all bounding boxes have probability of 1.

[pderrenger]

Thank you for the detailed feedback. The issue with all bounding boxes having a probability of 1 and the excessive number of detections suggests a potential bug. Please ensure you are using the latest version of the ultralytics package. If the issue persists, try running the model on a different image or dataset to see if the problem is consistent. If it is, please open an issue on our GitHub repository with details about your environment and the steps to reproduce the issue. This will help us investigate and resolve the problem more effectively.