def eval_all_models(cfg_file=cfg_file, begin_at=4000, add_eval_flag=True):
cfg.merge_from_file(cfg_file)
cfg.freeze()
root = cfg.OUTPUT_DIR
last_ckpt_file = os.path.join(root, 'last_checkpoint')
step = cfg.SOLVER.CHECKPOINT_PERIOD
max_iter = cfg.SOLVER.MAX_ITER
with open(last_ckpt_file, 'r') as f:
last_ckpt = f.read()
last_ckpt_dirname = os.path.dirname(last_ckpt)
def evaluation(main):
def evaluation_all_models(add_eval_flag=add_eval_flag):
for i in range(begin_at, max_iter + step, step):
last_ckpt = os.path.join(last_ckpt_dirname,
'model_{:07}.pth'.format(i))
with open(last_ckpt_file, 'w') as f:
f.write(last_ckpt)
main(add_eval_flag)
torch.cuda.empty_cache()
print('All models testing finished...')
return evaluation_all_models
return evaluation
def main():
parser = argparse.ArgumentParser(
description="PyTorch Object Detection Training")
parser.add_argument(
"--config-file",
default=
"/home/lxl/jittor/detectron.jittor/configs/maskrcnn_benchmark/e2e_faster_rcnn_R_50_C4_1x.yaml",
metavar="FILE",
help="path to config file",
type=str,
)
parser.add_argument("--local_rank", type=int, default=0)
parser.add_argument(
"--skip-test",
dest="skip_test",
help="Do not test the final model",
action="store_true",
)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
num_gpus = int(
os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
output_dir = cfg.OUTPUT_DIR
if output_dir:
mkdir(output_dir)
logger = setup_logger("detectron", output_dir)
logger.info("Using {} GPUs".format(num_gpus))
logger.info("Collecting env info (might take some time)")
logger.info("\n" + collect_env_info())
logger.info("Loaded configuration file {}".format(args.config_file))
model = train(cfg, args.local_rank, False)
if not args.skip_test:
run_test(cfg, model, False)
def run_inference(config_file):
import jittor as jt
from jittor_utils import auto_diff
from detectron.config import cfg
from detectron.modeling.detector import build_detection_model
from detectron.utils.checkpoint import DetectronCheckpointer
from detectron.data import make_data_loader
from detectron.engine.inference import inference
from detectron.utils.logger import setup_logger
jt.flags.use_cuda = 1
jt.cudnn.set_algorithm_cache_size(0)
cfg.merge_from_file(config_file)
cfg.freeze()
save_dir = ""
logger = setup_logger("maskrcnn_benchmark", save_dir)
model = build_detection_model(cfg)
# hook = auto_diff.Hook('fasterrcnn')
# hook.hook_module(model)
output_dir = cfg.OUTPUT_DIR
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
_ = checkpointer.load(cfg.MODEL.WEIGHT)
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
if cfg.MODEL.KEYPOINT_ON:
iou_types = iou_types + ("keypoints", )
dataset_names = cfg.DATASETS.TEST
data_loaders_val = make_data_loader(cfg, is_train=False)
for dataset_name, data_loader_val in zip(dataset_names, data_loaders_val):
inference(
model,
data_loader_val,
dataset_name=dataset_name,
iou_types=iou_types,
box_only=False if cfg.MODEL.FCOS_ON or cfg.MODEL.EMBED_MASK_ON
or cfg.MODEL.RETINANET_ON else cfg.MODEL.RPN_ONLY,
bbox_aug=cfg.TEST.BBOX_AUG.ENABLED,
expected_results=cfg.TEST.EXPECTED_RESULTS,
expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL,
cfg=cfg)
def run_torch_inference(config_file):
import jittor as jt
from jittor_utils import auto_diff
from maskrcnn_benchmark.config import cfg
from maskrcnn_benchmark.modeling.detector import build_detection_model
from maskrcnn_benchmark.utils.checkpoint import DetectronCheckpointer
from maskrcnn_benchmark.data import make_data_loader
from maskrcnn_benchmark.engine.inference import inference
from maskrcnn_benchmark.utils.logger import setup_logger
cfg.merge_from_file(config_file)
cfg.freeze()
save_dir = ""
model = build_detection_model(cfg)
model = model.cuda()
# hook = auto_diff.Hook('fasterrcnn')
# hook.hook_module(model)
output_dir = cfg.OUTPUT_DIR
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
_ = checkpointer.load(cfg.MODEL.WEIGHT)
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
dataset_names = cfg.DATASETS.TEST
data_loaders_val = make_data_loader(cfg, is_train=False)
for dataset_name, data_loader_val in zip(dataset_names, data_loaders_val):
inference(
model,
data_loader_val,
'coco_2014_minival',
iou_types=iou_types,
box_only=cfg.MODEL.RPN_ONLY,
expected_results=cfg.TEST.EXPECTED_RESULTS,
expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL,
)
# Given an url of an image, downloads the image and
# returns a PIL image
# """
# response = requests.get(url)
# pil_image = Image.open(BytesIO(response.content)).convert("RGB")
# # convert to BGR format
# image = np.array(pil_image)[:, :, [2, 1, 0]]
# return image
# turn on cuda
jt.flags.use_cuda = 1
# set config
config_file = '../configs/maskrcnn_benchmark/e2e_mask_rcnn_R_50_FPN_1x.yaml'
# update the config options with the config file
cfg.merge_from_file(config_file)
#cfg.MODEL.WEIGHT = "weight/maskrcnn_r50.pth"
#set predictor
coco_demo = COCODemo(
cfg,
min_image_size=800,
confidence_threshold=0.5,
)
#load image
pil_image = Image.open('test.jpg').convert("RGB")
image = np.array(pil_image)[:, :, [2, 1, 0]]
# compute predictions
predictions = coco_demo.run_on_opencv_image(image)
def main(cfg_file):
parser = argparse.ArgumentParser(
description="PyTorch Object Detection Training")
parser.add_argument(
"--config-file",
default=cfg_file,
metavar="FILE",
help="path to config file",
type=str,
)
parser.add_argument("--local_rank",
type=int,
default=int(
os.environ['CUDA_VISIBLE_DIVICES'])) # default=0
parser.add_argument(
"--skip-test",
dest="skip_test",
help="Do not test the final model",
action="store_true",
default=False, # True False
)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
num_gpus = int(
os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
args.distributed = num_gpus > 1
# if args.distributed:
# torch.cuda.set_device(args.local_rank)
# torch.distributed.init_process_group(
# backend="nccl", init_method="env://"
# )
# synchronize()
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
output_dir = cfg.OUTPUT_DIR
if output_dir:
mkdir(output_dir)
logger = setup_logger("detectron", output_dir, get_rank())
logger.info("Using {} GPUs".format(num_gpus))
logger.info(args)
logger.info("Loaded configuration file {}".format(args.config_file))
with open(args.config_file, "r") as cf:
config_str = "\n" + cf.read()
logger.info(config_str)
logger.info("Running with config:\n{}".format(cfg))
model = train(cfg, args.distributed)
if not args.skip_test:
run_test(cfg, model, args.distributed)
def run_model(config_file, img_f=None):
original_image = load(img_f)
from detectron.config import cfg
from detectron.modeling.detector import build_detection_model
from detectron.utils.checkpoint import DetectronCheckpointer
from detectron.structures.image_list import to_image_list
from detectron.modeling.roi_heads.mask_head.inference import Masker
from jittor import transform as T
from jittor import nn
import jittor as jt
from jittor_utils import auto_diff
jt.flags.use_cuda = 1
confidence_threshold = 0.0
cfg.merge_from_file(config_file)
model = build_detection_model(cfg)
checkpointer = DetectronCheckpointer(cfg, model, save_dir=cfg.OUTPUT_DIR)
_ = checkpointer.load(cfg.MODEL.WEIGHT)
name = config_file.split('/')[-1].split('.')[0]
# hook = auto_diff.Hook(name)
# hook.hook_module(model)
model.eval()
class Resize(object):
def __init__(self, min_size, max_size):
self.min_size = min_size
self.max_size = max_size
# modified from torchvision to add support for max size
def get_size(self, image_size):
w, h = image_size
size = self.min_size
max_size = self.max_size
if max_size is not None:
min_original_size = float(min((w, h)))
max_original_size = float(max((w, h)))
if max_original_size / min_original_size * size > max_size:
size = int(
round(max_size * min_original_size /
max_original_size))
if (w <= h and w == size) or (h <= w and h == size):
return (h, w)
if w < h:
ow = size
oh = int(size * h / w)
else:
oh = size
ow = int(size * w / h)
return (oh, ow)
def __call__(self, image):
size = self.get_size(image.size)
image = T.resize(image, size)
return image
def build_transform():
if cfg.INPUT.TO_BGR255:
to_bgr_transform = T.Lambda(lambda x: x * 255)
else:
to_bgr_transform = T.Lambda(lambda x: x[[2, 1, 0]])
normalize_transform = T.ImageNormalize(mean=cfg.INPUT.PIXEL_MEAN,
std=cfg.INPUT.PIXEL_STD)
min_size = cfg.INPUT.MIN_SIZE_TEST
max_size = cfg.INPUT.MAX_SIZE_TEST
transform = T.Compose([
T.ToPILImage(),
Resize(min_size, max_size),
T.ToTensor(),
to_bgr_transform,
normalize_transform,
])
return transform
transforms = build_transform()
image = transforms(original_image)
image_list = to_image_list(image, cfg.DATALOADER.SIZE_DIVISIBILITY)
predictions = model(image_list)
predictions = predictions[0]
if predictions.has_field("mask_scores"):
scores = predictions.get_field("mask_scores")
else:
scores = predictions.get_field("scores")
keep = jt.nonzero(scores > confidence_threshold).squeeze(1)
predictions = predictions[keep]
scores = predictions.get_field("scores")
idx, _ = jt.argsort(scores, 0, descending=True)
predictions = predictions[idx]
result_diff(predictions)
def main():
jt.flags.use_cuda = 1
parent_path = os.path.abspath(__file__).split("/tools/")[0]
parser = argparse.ArgumentParser(description="Object Detection Inference")
parser.add_argument(
"--config-file",
default=
f"{parent_path}/configs/maskrcnn_benchmark/e2e_mask_rcnn_R_50_FPN_1x.yaml",
metavar="FILE",
help="path to config file",
)
parser.add_argument(
"--ckpt",
help=
"The path to the checkpoint for test, default is the latest checkpoint.",
default=None,
)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
save_dir = ""
logger = setup_logger("detectron", save_dir)
logger.info("Using {} GPUs".format(1))
logger.info(cfg)
logger.info("Collecting env info (might take some time)")
logger.info("\n" + collect_env_info())
model = build_detection_model(cfg)
output_dir = cfg.OUTPUT_DIR
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
ckpt = cfg.MODEL.WEIGHT if args.ckpt is None else args.ckpt
_ = checkpointer.load(ckpt, use_latest=args.ckpt is None)
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
if cfg.MODEL.KEYPOINT_ON:
iou_types = iou_types + ("keypoints", )
output_folders = [None] * len(cfg.DATASETS.TEST)
dataset_names = cfg.DATASETS.TEST
if cfg.OUTPUT_DIR:
for idx, dataset_name in enumerate(dataset_names):
output_folder = os.path.join(cfg.OUTPUT_DIR, "inference",
dataset_name)
mkdir(output_folder)
output_folders[idx] = output_folder
data_loaders_val = make_data_loader(cfg, is_train=False)
for output_folder, dataset_name, data_loader_val in zip(
output_folders, dataset_names, data_loaders_val):
inference(
model,
data_loader_val,
dataset_name=dataset_name,
iou_types=iou_types,
box_only=False if cfg.MODEL.RETINANET_ON else cfg.MODEL.RPN_ONLY,
bbox_aug=cfg.TEST.BBOX_AUG.ENABLED,
expected_results=cfg.TEST.EXPECTED_RESULTS,
expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL,
output_folder=output_folder,
)
def main(add_eval_flag=False):
parser = argparse.ArgumentParser(
description="PyTorch Object Detection Inference")
parser.add_argument(
"--config-file",
default=cfg_file,
metavar="FILE",
help="path to config file",
)
parser.add_argument("--local_rank",
type=int,
default=int(os.environ['CUDA_VISIBLE_DIVICES']))
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
num_gpus = int(
os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1
args.distributed = num_gpus > 1
# if args.distributed:
# torch.cuda.set_device(args.local_rank)
# torch.distributed.init_process_group(
# backend="nccl", init_method="env://"
# )
# synchronize()
cfg.merge_from_file(args.config_file)
cfg.merge_from_list(args.opts)
cfg.freeze()
save_dir = ""
logger = setup_logger("detectron", save_dir, get_rank())
logger.info("Using {} GPUs".format(num_gpus))
logger.info(cfg)
model = build_detection_model(cfg)
model.to(cfg.MODEL.DEVICE)
output_dir = cfg.OUTPUT_DIR
checkpointer = DetectronCheckpointer(cfg, model, save_dir=output_dir)
_ = checkpointer.load(cfg.MODEL.WEIGHT)
iou_types = ("bbox", )
if cfg.MODEL.MASK_ON:
iou_types = iou_types + ("segm", )
output_folders = [None] * len(cfg.DATASETS.TEST)
dataset_names = cfg.DATASETS.TEST
if cfg.OUTPUT_DIR:
for idx, dataset_name in enumerate(dataset_names):
output_folder = os.path.join(cfg.OUTPUT_DIR, "inference",
dataset_name)
mkdir(output_folder)
output_folders[idx] = output_folder
data_loaders_val = make_data_loader(cfg,
is_train=False,
is_distributed=args.distributed)
for output_folder, dataset_name, data_loader_val in zip(
output_folders, dataset_names, data_loaders_val):
coco_results, _ = \
inference(
model,
data_loader_val,
dataset_name=dataset_name,
iou_types=iou_types,
box_only=cfg.MODEL.RPN_ONLY,
device=cfg.MODEL.DEVICE,
expected_results=cfg.TEST.EXPECTED_RESULTS,
expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL,
output_folder=output_folder,
)
synchronize()
def add_eval_fields():
ar = coco_results.results['bbox']['AR50']
ap = coco_results.results['bbox']['AP50']
checkpoint_file = checkpointer.get_checkpoint_file()
base_checkpoint_file = os.path.basename(checkpoint_file).split('.')[0]
new_checkpoint_file = os.path.join(
output_dir,
base_checkpoint_file + '_ar{:.03}_ap_{:.03}.pth'.format(ar, ap))
os.rename(checkpoint_file, new_checkpoint_file)
if add_eval_flag:
add_eval_fields()