I am very new to this repository, and in-fact to machine learning. So I highly appreciate your time and help.
I am following the link called Introduction to SAHI which is given by https://medium.com/codable/sahi-a-vision-library-for-performing-sliced-inference-on-large-images-small-objects-c8b086af3b80
What I did was just copy the code snippets given and paste them in my Spyder environment.
The final code I have is the following:
`from sahi import AutoDetectionModel
from sahi.predict import get_sliced_prediction
from sahi.utils.cv import read_image_as_pil
from pathlib import Path
from IPython.display import Image, display
In the error above, the mmdetection-main is the repository of mmdetection.
Note: All the pip are apparently installed in site-packages given in the above directory. So I downloaded and copied the mmdetection repository also in the same directory.
In case it's relevant, the system I am using is the following:
OS: Windows 11
Processor: Core i5
Graphics Card: Radeon 2GB (I don't have Nvidia)
RAM: 8GB
Environment: Anaconda Spyder
I am very new to this repository, and in-fact to machine learning. So I highly appreciate your time and help. I am following the link called Introduction to SAHI which is given by https://medium.com/codable/sahi-a-vision-library-for-performing-sliced-inference-on-large-images-small-objects-c8b086af3b80 What I did was just copy the code snippets given and paste them in my Spyder environment. The final code I have is the following: `from sahi import AutoDetectionModel from sahi.predict import get_sliced_prediction from sahi.utils.cv import read_image_as_pil from pathlib import Path from IPython.display import Image, display
detection_model = AutoDetectionModel.from_pretrained( model_type='mmdet', model_path="C:/Users/abdul/AppData/Roaming/Python/Python311/site-packages/mmdetection-main/mmdet", config_path="C:/Users/abdul/AppData/Roaming/Python/Python311/site-packages/mmdetection-main/mmdet", confidence_threshold=0.4, device="cuda:0" )
print(Path.cwd()) #It shows the current working directory
image = read_image_as_pil("cars_on_road.png")
result = get_sliced_prediction( image, detection_model, slice_height = 256, slice_width = 256, overlap_height_ratio = 0.2, overlap_width_ratio = 0.2 )
result.export_visuals(export_dir="result/")
result_image_path = "result/prediction_visual.png" display(Image(result_image_path))`
After running it, I am getting the following error:
The things I have done:
pip install mmdet mmcv-full
In the error above, the mmdetection-main is the repository of mmdetection.
Note: All the pip are apparently installed in site-packages given in the above directory. So I downloaded and copied the mmdetection repository also in the same directory.
In case it's relevant, the system I am using is the following: OS: Windows 11 Processor: Core i5 Graphics Card: Radeon 2GB (I don't have Nvidia) RAM: 8GB Environment: Anaconda Spyder