xclele/yolov10
1
0
Fork 0
mirror of https://github.com/THU-MIG/yolov10.git synced 2025-05-23 21:44:22 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

fixed run_val

Browse Source
This commit is contained in:
nielseni6 2024-10-15 11:53:27 -04:00
parent b02bf58de6
commit 2a95a652bd
1 changed files with 17 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

58
run_val.py
View File

@ -1,51 +1,27 @@
from ultralytics import YOLOv10 from ultralytics import YOLOv10, YOLO
import torch # from ultralytics.engine.pgt_trainer import PGTTrainer
from PIL import Image # from ultralytics import BaseTrainer
from torchvision import transforms # from ultralytics.engine.trainer import BaseTrainer
import os
# Define the device # Set CUDA device (only needed for multi-gpu machines)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "4"
# model = YOLOv10()
# model = YOLO()
# If you want to finetune the model with pretrained weights, you could load the
# pretrained weights like below
# model = YOLOv10.from_pretrained('jameslahm/yolov10{n/s/m/b/l/x}') # model = YOLOv10.from_pretrained('jameslahm/yolov10{n/s/m/b/l/x}')
# model = YOLOv10.from_pretrained('jameslahm/yolov10n')
# or # or
# wget https://github.com/THU-MIG/yolov10/releases/download/v1.1/yolov10{n/s/m/b/l/x}.pt # wget https://github.com/THU-MIG/yolov10/releases/download/v1.1/yolov10{n/s/m/b/l/x}.pt
# wget https://github.com/THU-MIG/yolov10/releases/download/v1.1/yolov10n.pt model = YOLOv10('yolov10n.pt')
# model = YOLOv10('yolov10{n/s/m/b/l/x}.pt')
model = YOLOv10('yolov10n.pt').to(device)
# Load the image # Evaluate the model on the validation set
# path = '/home/nielseni6/PythonScripts/Github/yolov10/images/fat-dog.jpg' results = model.val(data='coco.yaml')
path = '/home/nielseni6/PythonScripts/Github/yolov10/images/The-Cardinal-Bird.jpg'
image = Image.open(path)
# Define the transformation to resize the image, convert it to a tensor, and normalize it # Print the evaluation results
transform = transforms.Compose([ print(results)
transforms.Resize((640, 640)),
transforms.ToTensor(),
# transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
# Apply the transformation
image_tensor = transform(image)
# Add a batch dimension
image_tensor = image_tensor.unsqueeze(0).to(device)
image_tensor = image_tensor.requires_grad_(True)
# Predict for a specific image
# results = model.predict(image_tensor, save=True)
# model.requires_grad_(True)
# for p in model.parameters():
# p.requires_grad = True
results = model.predict(image_tensor, save=True)
# Display the results
for result in results:
print(result)
# pred = results[0].boxes[0].conf # pred = results[0].boxes[0].conf