github:https://github.com/icey-zhang/SuperYOLO
article:https://www.sfu.ca/~zhenman/files/J12-TGRS2023-SuperYOLO.pdf
环境:
PyTorch 2.5.1
Python 3.12(ubuntu22.04)
CUDA 12.4
GPU RTX 4090D(24GB) * 1升降配置
CPU 18 vCPU AMD EPYC 9754 128-Core Processor
内存 60GB
由于不习惯SuperYOLO的数据集读取方式,此处使用原YOLOv5方式。
此处使用的数据集为VisDrone2019。
大中小三种尺度划分:
Small
data.yaml内容示例:
train: /root/autodl-tmp/VisDrone/VisDrone2019-DET-train/images # train images (relative to 'path') 6471 images
val: /root/autodl-tmp/VisDrone/VisDrone2019-DET-val/images # val images (relative to 'path') 548 images
test: /root/autodl-tmp/VisDrone/VisDrone2019-DET-test-dev/images # test images (optional) 1610 images
# Classes
nc: 10
names: ['pedestrian', 'people', 'bicycle', 'car', 'van', 'truck', 'tricycle', 'awning-tricycle', 'bus', 'motor']
相应的改动:
utils/general.py:
# 替换原check_dataset()函数
def check_dataset(data, autodownload=True):
# Download (optional)
extract_dir = ''
if isinstance(data, (str, Path)) and str(data).endswith('.zip'): # i.e. gs://bucket/dir/coco128.zip
download(data, dir='../datasets', unzip=True, delete=False, curl=False, threads=1)
data = next((Path('../datasets') / Path(data).stem).rglob('*.yaml'))
extract_dir, autodownload = data.parent, False
# Read yaml (optional)
if isinstance(data, (str, Path)):
with open(data, errors='ignore') as f:
data = yaml.safe_load(f) # dictionary
# Parse yaml
path = extract_dir or Path(data.get('path') or '') # optional 'path' default to '.'
for k in 'train', 'val', 'test':
if data.get(k): # prepend path
data[k] = str(path / data[k]) if isinstance(data[k], str) else [str(path / x) for x in data[k]]
assert 'nc' in data, "Dataset 'nc' key missing."
if 'names' not in data:
data['names'] = [f'class{i}' for i in range(data['nc'])] # assign class names if missing
train, val, test, s = (data.get(x) for x in ('train', 'val', 'test', 'download'))
if val:
val = [Path(x).resolve() for x in (val if isinstance(val, list) else [val])] # val path
if not all(x.exists() for x in val):
print('nWARNING: Dataset not found, nonexistent paths: %s' % [str(x) for x in val if not x.exists()])
if s and autodownload: # download script
root = path.parent if 'path' in data else '..' # unzip directory i.e. '../'
if s.startswith('http') and s.endswith('.zip'): # URL
f = Path(s).name # filename
print(f'Downloading {s} to {f}...')
torch.hub.download_url_to_file(s, f)
Path(root).mkdir(parents=True, exist_ok=True) # create root
ZipFile(f).extractall(path=root) # unzip
Path(f).unlink() # remove zip
r = None # success
elif s.startswith('bash '): # bash script
print(f'Running {s} ...')
r = os.system(s)
else: # python script
r = exec(s, {'yaml': data}) # return None
print(f"Dataset autodownload {f'success, saved to {root}' if r in (0, None) else 'failure'}n")
else:
raise Exception('Dataset not found.')
return data # dictionary
单RGB分支评估完整代码(test.py):
import argparse
import json
import os
from pathlib import Path
from threading import Thread
import numpy as np
import torch
import yaml
from tqdm import tqdm
from models.experimental import attempt_load
from utils.datasets import create_dataloader, create_dataloader_sr
from utils.general import coco80_to_coco91_class, check_dataset, check_file, check_img_size, check_requirements,
box_iou, non_max_suppression,weighted_boxes, scale_coords, xyxy2xywh, xywh2xyxy, set_logging, increment_path, colorstr
from utils.metrics import ap_per_class, ConfusionMatrix
from utils.plots import plot_images, output_to_target, plot_study_txt
from utils.torch_utils import select_device, time_synchronized
from torchvision import transforms
from PIL import Image
unloader = transforms.ToPILImage()
def tensor_to_PIL(tensor):
image = tensor.cpu().clone()
image = image.squeeze(0)
image = unloader(image)
image.save('a.png')
return image
def process_batch(detections, labels, iouv):
correct = torch.zeros(detections.shape[0], iouv.shape[0], dtype=torch.bool, device=iouv.device)
iou = box_iou(labels[:, 1:], detections[:, :4])
x = torch.where((iou >= iouv[0]) & (labels[:, 0:1] == detections[:, 5])) # IoU above threshold and classes match
if x[0].shape[0]:
matches = torch.cat((torch.stack(x, 1), iou[x[0], x[1]][:, None]), 1).cpu().numpy() # [label, detection, iou]
if x[0].shape[0] > 1:
matches = matches[matches[:, 2].argsort()[::-1]]
matches = matches[np.unique(matches[:, 1], return_index=True)[1]]
matches = matches[np.unique(matches[:, 0], return_index=True)[1]]
matches = torch.Tensor(matches).to(iouv.device)
correct[matches[:, 1].long()] = matches[:, 2:3] >= iouv
return correct
def test(data,
weights=None,
batch_size=32,
imgsz=640,
input_mode = None,
conf_thres=0.001,
iou_thres=0.6, # for NMS
save_json=False,
single_cls=False,
augment=False,
verbose=False,
model=None,
dataloader=None,
save_dir=Path(''), # for saving images
save_txt=False, # for auto-labelling
save_hybrid=False, # for hybrid auto-labelling
save_conf=False, # save auto-label confidences
plots=True,
wandb_logger=None,
compute_loss=None,
is_coco=False):
print(f"*******************当前权重:{weights}*******************")
# Initialize/load model and set device
training = model is not None
if training: # called by train.py
device = next(model.parameters()).device # get model device
else: # called directly
set_logging()
device = select_device(opt.device, batch_size=batch_size)
# Directories
save_dir = Path(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok)) # increment run
(save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir
# Load model
model = attempt_load(weights, map_location=device) # load FP32 model
print(model.yaml_file)
gs = max(int(model.stride.max()), 32) # grid size (max stride)
imgsz = check_img_size(imgsz, s=gs) # check img_size
data = check_dataset(data) # check
half = device.type != 'cpu'
model.half() if half else model.float()
# Configure
model.eval()
# print(model)
if isinstance(data, str):
is_coco = data.endswith('coco.yaml')
with open(data) as f:
data = yaml.load(f, Loader=yaml.SafeLoader)
check_dataset(data) # check
nc = 1 if single_cls else int(data['nc']) # number of classes
iouv = torch.linspace(0.5, 0.95, 10).to(device) # iou vector for mAP@0.5:0.95 zjq
niou = iouv.numel()
stats_small, stats_medium, stats_large = [], [], []
# Logging
log_imgs = 0
if wandb_logger and wandb_logger.wandb:
log_imgs = min(wandb_logger.log_imgs, 100)
if not training:
task = opt.task if opt.task in ('train', 'val', 'test') else 'val' # path to train/val/test images
if opt.data.endswith('vedai.yaml') or opt.data.endswith('SRvedai.yaml'):
from utils.datasets import create_dataloader_sr as create_dataloader
else:
from utils.datasets import create_dataloader
dataloader, dataset = create_dataloader(data[task], imgsz, batch_size, gs, opt,
augment=augment, cache=True,
rect=True,
quad=False, prefix=colorstr(f'{task}: '))
seen = 0
confusion_matrix = ConfusionMatrix(nc=nc)
names = {k: v for k, v in enumerate(model.names if hasattr(model, 'names') else model.module.names)}
coco91class = coco80_to_coco91_class()
s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Labels', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')
p, r, f1, mp, mr, map50, map, t0, t1 = 0., 0., 0., 0., 0., 0., 0., 0., 0.
loss = torch.zeros(3, device=device)
jdict, stats, ap, ap_class, wandb_images = [], [], [], [], []
for batch_i, (img, ir, targets, paths, shapes) in enumerate(tqdm(dataloader, desc=s)): #zjq
# t1 = time_sync()
img = img.to(device, non_blocking=True).float()
img = img.half() if half else img.float() # uint8 to fp16/32
img /= 255.0 # 0 - 255 to 0.0 - 1.0
ir = ir.to(device, non_blocking=True).float()
# ir = ir.half() if half else ir.float() # uint8 to fp16/32
ir /= 255.0 # 0 - 255 to 0.0 - 1.0
targets = targets.to(device)
nb, _, height, width = img.shape # batch size, channels, height, width
with torch.no_grad():
# Run model
t = time_synchronized()
try:
out, train_out = model(img,ir,input_mode=input_mode) #zjq inference and training outputs
except:
out, train_out,_ = model(img,ir,input_mode=input_mode) #zjq inference and training outputs
t0 += time_synchronized() - t
# Compute loss
if compute_loss:
loss += compute_loss([x.float() for x in train_out], targets)[1][:3] # box, obj, cls
# Run NMS
targets[:, 2:] *= torch.Tensor([width, height, width, height]).to(device) # to pixels
lb = [targets[targets[:, 0] == i, 1:] for i in range(nb)] if save_hybrid else [] # for autolabelling
t = time_synchronized()
out = non_max_suppression(out, conf_thres=conf_thres, iou_thres=iou_thres, labels=lb, multi_label=True)
# out = weighted_boxes(out,image_size=imgsz, conf_thres=conf_thres, iou_thres=iou_thres, labels=lb, multi_label=True)
t1 += time_synchronized() - t
# Statistics per image
for si, pred in enumerate(out):
labels = targets[targets[:, 0] == si, 1:]
nl = len(labels)
tcls = labels[:, 0].tolist() if nl else [] # target class
# path = Path(paths[si])
path, shape = Path(paths[si]), shapes[si][0]
seen += 1
if len(pred) == 0:
if nl:
stats.append((torch.zeros(0, niou, dtype=torch.bool), torch.Tensor(), torch.Tensor(), tcls))
# Calculate the xyxy format of the ground truth
tbox = xywh2xyxy(labels[:, 1:5])
scale_coords(img[si].shape[1:], tbox, shape, shapes[si][1])
gt_boxes = torch.cat((labels[:, 0:1], tbox), 1)
else:
stats.append((torch.zeros(0, niou, dtype=torch.bool), torch.Tensor(), torch.Tensor(), []))
gt_boxes = torch.empty((0, 5), device=img.device)
# Divide ground truth by scale
small_thresh = 32 * 32
medium_thresh = 96 * 96
if gt_boxes.shape[0] > 0:
gt_areas = (gt_boxes[:, 3] - gt_boxes[:, 1]) * (gt_boxes[:, 4] - gt_boxes[:, 2])
small_gt_idx = gt_areas = small_thresh) & (gt_areas = medium_thresh
small_gts = gt_boxes[small_gt_idx]
medium_gts = gt_boxes[medium_gt_idx]
large_gts = gt_boxes[large_gt_idx]
else:
small_gts = medium_gts = large_gts = torch.empty((0, 5), device=img.device)
stats_small.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(),
small_gts[:, 0].cpu() if small_gts.shape[0] > 0 else torch.Tensor()))
stats_medium.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(),
medium_gts[:, 0].cpu() if medium_gts.shape[0] > 0 else torch.Tensor()))
stats_large.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(),
large_gts[:, 0].cpu() if large_gts.shape[0] > 0 else torch.Tensor()))
continue
# If there are prediction results, convert them to the original image scale
if single_cls:
pred[:, 5] = 0
# Predictions
predn = pred.clone()
scale_coords(img[si].shape[1:], predn[:, :4], shape, shapes[si][1]) # native-space pred
if nl:
tbox = xywh2xyxy(labels[:, 1:5])
scale_coords(img[si].shape[1:], tbox, shape, shapes[si][1])
labelsn = torch.cat((labels[:, 0:1], tbox), 1)
correct = process_batch(predn, labelsn, iouv)
if plots:
confusion_matrix.process_batch(predn, labelsn)
else:
correct = torch.zeros(pred.shape[0], niou, dtype=torch.bool)
stats.append((correct.cpu(), pred[:, 4].cpu(), pred[:, 5].cpu(), tcls))
# Calculate the statistical information of scale grouping
if nl:
gt_boxes = torch.cat((labels[:, 0:1], tbox), 1)
else:
gt_boxes = torch.empty((0, 5), device=img.device)
def compute_area(boxes):
return (boxes[:, 2] - boxes[:, 0]) * (boxes[:, 3] - boxes[:, 1])
if predn.shape[0] > 0:
pred_areas = compute_area(predn[:, :4])
else:
pred_areas = torch.tensor([], device=img.device)
if gt_boxes.shape[0] > 0:
gt_areas = compute_area(gt_boxes[:, 1:5])
else:
gt_areas = torch.tensor([], device=img.device)
small_thresh = 32 * 32
medium_thresh = 96 * 96
if pred_areas.numel() > 0:
small_pred_idx = pred_areas = small_thresh) & (pred_areas = medium_thresh
small_preds = predn[small_pred_idx]
medium_preds = predn[medium_pred_idx]
large_preds = predn[large_pred_idx]
else:
small_preds = medium_preds = large_preds = torch.empty((0, 6), device=img.device)
if gt_boxes.shape[0] > 0:
small_gt_idx = gt_areas = small_thresh) & (gt_areas = medium_thresh
small_gts = gt_boxes[small_gt_idx]
medium_gts = gt_boxes[medium_gt_idx]
large_gts = gt_boxes[large_gt_idx]
else:
small_gts = medium_gts = large_gts = torch.empty((0, 5), device=img.device)
# Small
if small_preds.shape[0] == 0 and small_gts.shape[0] == 0:
stats_small.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(), torch.Tensor()))
else:
if small_preds.shape[0] > 0 and small_gts.shape[0] > 0:
correct_small = process_batch(small_preds, small_gts, iouv)
else:
correct_small = torch.zeros(small_preds.shape[0], iouv.shape[0], dtype=torch.bool, device=iouv.device) if small_preds.shape[0] > 0 else torch.zeros(0, iouv.shape[0], dtype=torch.bool, device=iouv.device)
stats_small.append((correct_small.cpu(),
small_preds[:, 4].cpu() if small_preds.shape[0] > 0 else torch.Tensor(),
small_preds[:, 5].cpu() if small_preds.shape[0] > 0 else torch.Tensor(),
small_gts[:, 0].cpu() if small_gts.shape[0] > 0 else torch.Tensor()))
# Medium
if medium_preds.shape[0] == 0 and medium_gts.shape[0] == 0:
stats_medium.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(), torch.Tensor()))
else:
if medium_preds.shape[0] > 0 and medium_gts.shape[0] > 0:
correct_medium = process_batch(medium_preds, medium_gts, iouv)
else:
correct_medium = torch.zeros(medium_preds.shape[0], iouv.shape[0], dtype=torch.bool, device=iouv.device) if medium_preds.shape[0] > 0 else torch.zeros(0, iouv.shape[0], dtype=torch.bool, device=iouv.device)
stats_medium.append((correct_medium.cpu(),
medium_preds[:, 4].cpu() if medium_preds.shape[0] > 0 else torch.Tensor(),
medium_preds[:, 5].cpu() if medium_preds.shape[0] > 0 else torch.Tensor(),
medium_gts[:, 0].cpu() if medium_gts.shape[0] > 0 else torch.Tensor()))
# Large
if large_preds.shape[0] == 0 and large_gts.shape[0] == 0:
stats_large.append((torch.zeros(0, iouv.shape[0], dtype=torch.bool),
torch.Tensor(), torch.Tensor(), torch.Tensor()))
else:
if large_preds.shape[0] > 0 and large_gts.shape[0] > 0:
correct_large = process_batch(large_preds, large_gts, iouv)
else:
correct_large = torch.zeros(large_preds.shape[0], iouv.shape[0], dtype=torch.bool, device=iouv.device) if large_preds.shape[0] > 0 else torch.zeros(0, iouv.shape[0], dtype=torch.bool, device=iouv.device)
stats_large.append((correct_large.cpu(),
large_preds[:, 4].cpu() if large_preds.shape[0] > 0 else torch.Tensor(),
large_preds[:, 5].cpu() if large_preds.shape[0] > 0 else torch.Tensor(),
large_gts[:, 0].cpu() if large_gts.shape[0] > 0 else torch.Tensor()))
# end for each image in batch
# Append to text file
if save_txt:
gn = torch.tensor(shapes[si][0])[[1, 0, 1, 0]] # normalization gain whwh
for *xyxy, conf, cls in predn.tolist():
xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh
line = (cls, *xywh, conf) if save_conf else (cls, *xywh) # label format
with open(save_dir / 'labels' / (path.stem + '.txt'), 'a') as f:
f.write(('%g ' * len(line)).rstrip() % line + 'n')
# W&B logging - Media Panel Plots
if len(wandb_images) 0: # Check for test operation
if wandb_logger.current_epoch % wandb_logger.bbox_interval == 0:
box_data = [{"position": {"minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]},
"class_id": int(cls),
"box_caption": "%s %.3f" % (names[cls], conf),
"scores": {"class_score": conf},
"domain": "pixel"} for *xyxy, conf, cls in pred.tolist()]
boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space
wandb_images.append(wandb_logger.wandb.Image(img[si], boxes=boxes, caption=path.name))
wandb_logger.log_training_progress(predn, path, names) if wandb_logger and wandb_logger.wandb_run else None
# Append to pycocotools JSON dictionary
if save_json:
# [{"image_id": 42, "category_id": 18, "bbox": [258.15, 41.29, 348.26, 243.78], "score": 0.236}, ...
image_id = int(path.stem) if path.stem.isnumeric() else path.stem
box = xyxy2xywh(predn[:, :4]) # xywh
box[:, :2] -= box[:, 2:] / 2 # xy center to top-left corner
for p, b in zip(pred.tolist(), box.tolist()):
jdict.append({'image_id': image_id,
'category_id': coco91class[int(p[5])] if is_coco else int(p[5]),
'bbox': [round(x, 3) for x in b],
'score': round(p[4], 5)})
# Plot images
if plots: #and batch_i 0:
stats_small_agg = [np.concatenate(x, 0) for x in zip(*stats_small)]
p_small, r_small, ap_small, f1_small, ap_class_small = ap_per_class(*stats_small_agg, plot=False, names=names)
mAP_small = ap_small[:, 0].mean()
else:
mAP_small = 0.0
if len(stats_medium) and np.concatenate(list(zip(*stats_medium))[0]).size > 0:
stats_medium_agg = [np.concatenate(x, 0) for x in zip(*stats_medium)]
p_medium, r_medium, ap_medium, f1_medium, ap_class_medium = ap_per_class(*stats_medium_agg, plot=False, names=names)
mAP_medium = ap_medium[:, 0].mean()
else:
mAP_medium = 0.0
if len(stats_large) and np.concatenate(list(zip(*stats_large))[0]).size > 0:
stats_large_agg = [np.concatenate(x, 0) for x in zip(*stats_large)]
p_large, r_large, ap_large, f1_large, ap_class_large = ap_per_class(*stats_large_agg, plot=False, names=names)
mAP_large = ap_large[:, 0].mean()
else:
mAP_large = 0.0
# Print results
pf = '%20s' + '%12i' * 2 + '%12.4g' * 4 # print format
print(pf % ('all', seen, nt.sum(), mp, mr, map50, map))
# with open("trying.txt", 'a+') as f:
# f.write((pf % ('all', seen, nt.sum(), mp, mr, map50, map)) + 'n') # append metrics, val_loss
# Print results per class
if (verbose or (nc 1 and len(stats):
for i, c in enumerate(ap_class):
print(pf % (names[c], seen, nt[c], p[i], r[i], ap50[i], ap[i]))
# Print s-m-l Scale mAP
print("nScale-based mAP results:")
print(f" Small objects mAP: {mAP_small:.4f}")
print(f" Medium objects mAP: {mAP_medium:.4f}")
print(f" Large objects mAP: {mAP_large:.4f}")
# Print speeds
t = tuple(x / seen * 1E3 for x in (t0, t1, t0 + t1)) + (imgsz, imgsz, batch_size) # tuple
if not training:
print('Speed: %.3f/%.3f/%.3f ms inference/NMS/total per %gx%g image at batch-size %g' % t)
fps = 1000 / (t[0] + t[1] + t[2])
print(f'fps = {fps:.2f}')
# Plots
if plots:
confusion_matrix.plot(save_dir=save_dir, names=list(names.values()))
if wandb_logger and wandb_logger.wandb:
val_batches = [wandb_logger.wandb.Image(str(f), caption=f.name) for f in sorted(save_dir.glob('test*.jpg'))]
wandb_logger.log({"Validation": val_batches})
if wandb_images:
wandb_logger.log({"Bounding Box Debugger/Images": wandb_images})
# Save JSON
if save_json and len(jdict):
w = Path(weights[0] if isinstance(weights, list) else weights).stem if weights is not None else '' # weights
anno_json = '../coco/annotations/instances_val2017.json' # annotations json
pred_json = str(save_dir / f"{w}_predictions.json") # predictions json
print('nEvaluating pycocotools mAP... saving %s...' % pred_json)
with open(pred_json, 'w') as f:
json.dump(jdict, f)
try: # https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocoEvalDemo.ipynb
from pycocotools.coco import COCO
from pycocotools.cocoeval import COCOeval
anno = COCO(anno_json) # init annotations api
pred = anno.loadRes(pred_json) # init predictions api
eval = COCOeval(anno, pred, 'bbox')
if is_coco:
eval.params.imgIds = [int(Path(x).stem) for x in dataloader.dataset.img_files] # image IDs to evaluate
eval.evaluate()
eval.accumulate()
eval.summarize()
map, map50 = eval.stats[:2] # update results (mAP@0.5:0.95, mAP@0.5)
except Exception as e:
print(f'pycocotools unable to run: {e}')
# Return results
model.float() # for training
if not training:
s = f"n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
print(f"Results saved to {save_dir}{s}")
maps = np.zeros(nc) + map
for i, c in enumerate(ap_class):
maps[c] = ap[i]
return (mp, mr, map50, map, *(loss.cpu() / len(dataloader)).tolist()), maps, t
if __name__ == '__main__':
parser = argparse.ArgumentParser(prog='test.py')
parser.add_argument('--weights', nargs='+', type=str, default='/root/SuperYOLO-main/runs/train/exp20/weights/best.pt', help='model.pt path(s)')
parser.add_argument('--data', type=str, default='/root/SuperYOLO-main/visDrone-Copy1.yaml', help='*.data path')
parser.add_argument('--batch-size', type=int, default=8, help='size of each image batch')
parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)')
parser.add_argument('--input_mode', type=str, default='RGB') #RGB IR RGB+IR RGB+IR+fusion
parser.add_argument('--conf-thres', type=float, default=0.001, help='object confidence threshold')
parser.add_argument('--iou-thres', type=float, default=0.6, help='IOU threshold for NMS')
parser.add_argument('--task', default='val', help='train, val, test, speed or study')
parser.add_argument('--device', default='0', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--single-cls', action='store_true', help='treat as single-class dataset')
parser.add_argument('--augment', action='store_true', help='augmented inference')
parser.add_argument('--verbose', action='store_true', help='report mAP by class')
parser.add_argument('--save-txt', action='store_true', help='save results to *.txt')
parser.add_argument('--save-hybrid', action='store_true', help='save label+prediction hybrid results to *.txt')
parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels')
parser.add_argument('--save-json', action='store_true', help='save a cocoapi-compatible JSON results file')
parser.add_argument('--project', default='runs/test', help='save to project/name')
parser.add_argument('--name', default='exp', help='save to project/name')
parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
opt = parser.parse_args()
opt.save_json |= opt.data.endswith('coco.yaml')
opt.data = check_file(opt.data) # check file
print(opt)
check_requirements()
if opt.task in ('train', 'val', 'test'): # run normally
test(opt.data,
opt.weights,
opt.batch_size,
opt.img_size,
opt.input_mode,
opt.conf_thres,
opt.iou_thres,
opt.save_json,
opt.single_cls,
opt.augment,
opt.verbose,
save_txt=opt.save_txt | opt.save_hybrid,
save_hybrid=opt.save_hybrid,
save_conf=opt.save_conf,
)
elif opt.task == 'speed': # speed benchmarks
for w in opt.weights:
test(opt.data, w, opt.batch_size, opt.img_size, 0.25, 0.45, save_json=False, plots=False)
elif opt.task == 'study': # run over a range of settings and save/plot
# python test.py --task study --data coco.yaml --iou 0.7 --weights yolov5s.pt yolov5m.pt yolov5l.pt yolov5x.pt
x = list(range(256, 1536 + 128, 128)) # x axis (image sizes)
for w in opt.weights:
f = f'study_{Path(opt.data).stem}_{Path(w).stem}.txt' # filename to save to
y = [] # y axis
for i in x: # img-size
print(f'nRunning {f} point {i}...')
r, _, t = test(opt.data, w, opt.batch_size, i, opt.conf_thres, opt.iou_thres, opt.save_json,
plots=False)
y.append(r + t) # results and times
np.savetxt(f, y, fmt='%10.4g') # save
os.system('zip -r study.zip study_*.txt')
plot_study_txt(x=x) # plot
碎碎念:
作为以YOLO为baseline改进而来的模型,如果是初次使用SuperYOLO,想必会遇到各种奇怪的问题(
相当一部分是因为长期未得到更新导致的,其中不少报错可以在YOLOv5官方github的issue中找到解决方法。
至于剩下的,在我眼中可称为“开门劝退”的,就是那奇怪的数据集加载方式。个人的建议是,如果像我一样只需要使用RGB分支,还是想办法datasets.py中的代码调整回YOLOv5的经典形式最为省事,可以回避掉许多奇奇怪怪的问题。
当然,就效果而言,SuperYOLO还是相当不错的,即使不使用“RGB+IR”的双分支结构(VisDrone2019数据集)
参与评论
手机查看
返回顶部