108 lines
3.6 KiB
Python
Executable File
108 lines
3.6 KiB
Python
Executable File
import os
|
|
import onnx
|
|
import paddle
|
|
import paddle.nn as nn
|
|
import paddle.nn.functional as F
|
|
|
|
from ppdet.engine import Trainer
|
|
from ppdet.utils.cli import ArgsParser
|
|
from ppdet.utils.check import check_version, check_config
|
|
from ppdet.core.workspace import load_config, merge_config
|
|
|
|
|
|
class DeepStreamOutput(nn.Layer):
|
|
def __init__(self, img_size, use_focal_loss):
|
|
super().__init__()
|
|
self.img_size = img_size
|
|
self.use_focal_loss = use_focal_loss
|
|
|
|
def forward(self, x):
|
|
boxes = x['bbox']
|
|
convert_matrix = paddle.to_tensor(
|
|
[[1, 0, 1, 0], [0, 1, 0, 1], [-0.5, 0, 0.5, 0], [0, -0.5, 0, 0.5]], dtype=boxes.dtype
|
|
)
|
|
boxes @= convert_matrix
|
|
boxes *= paddle.to_tensor([[*self.img_size]]).flip(1).tile([1, 2]).unsqueeze(1)
|
|
bbox_num = F.sigmoid(x['bbox_num']) if self.use_focal_loss else F.softmax(x['bbox_num'])[:, :, :-1]
|
|
scores = paddle.max(bbox_num, axis=-1, keepdim=True)
|
|
labels = paddle.argmax(bbox_num, axis=-1, keepdim=True)
|
|
return paddle.concat((boxes, scores, paddle.cast(labels, dtype=boxes.dtype)), axis=-1)
|
|
|
|
|
|
def rtdetr_paddle_export(FLAGS):
|
|
cfg = load_config(FLAGS.config)
|
|
FLAGS.opt['weights'] = FLAGS.weights
|
|
FLAGS.opt['exclude_nms'] = True
|
|
FLAGS.opt['exclude_post_process'] = True
|
|
merge_config(FLAGS.opt)
|
|
merge_config(FLAGS.opt)
|
|
check_config(cfg)
|
|
check_version()
|
|
trainer = Trainer(cfg, mode='test')
|
|
trainer.load_weights(cfg.weights)
|
|
trainer.model.eval()
|
|
if not os.path.exists('.tmp'):
|
|
os.makedirs('.tmp')
|
|
static_model, _ = trainer._get_infer_cfg_and_input_spec('.tmp')
|
|
os.system('rm -r .tmp')
|
|
return trainer.cfg, static_model
|
|
|
|
|
|
def suppress_warnings():
|
|
import warnings
|
|
warnings.filterwarnings('ignore')
|
|
|
|
|
|
def main(FLAGS):
|
|
suppress_warnings()
|
|
|
|
print(f'\nStarting: {FLAGS.weights}')
|
|
|
|
print('Opening RT-DETR Paddle model')
|
|
|
|
paddle.set_device('cpu')
|
|
cfg, model = rtdetr_paddle_export(FLAGS)
|
|
|
|
img_size = [cfg.eval_size[1], cfg.eval_size[0]]
|
|
|
|
model = nn.Sequential(model, DeepStreamOutput(img_size, cfg.use_focal_loss))
|
|
|
|
onnx_input_im = {}
|
|
onnx_input_im['image'] = paddle.static.InputSpec(shape=[FLAGS.batch, 3, *img_size], dtype='float32')
|
|
onnx_output_file = f'{FLAGS.weights}.onnx'
|
|
|
|
print('Exporting the model to ONNX\n')
|
|
paddle.onnx.export(model, FLAGS.weights, input_spec=[onnx_input_im], opset_version=FLAGS.opset)
|
|
|
|
if FLAGS.simplify:
|
|
print('Simplifying the ONNX model')
|
|
import onnxslim
|
|
model_onnx = onnx.load(onnx_output_file)
|
|
model_onnx = onnxslim.slim(model_onnx)
|
|
onnx.save(model_onnx, onnx_output_file)
|
|
|
|
print(f'Done: {onnx_output_file}\n')
|
|
|
|
|
|
def parse_args():
|
|
parser = ArgsParser()
|
|
parser.add_argument('-w', '--weights', required=True, help='Input weights (.pdparams) file path (required)')
|
|
parser.add_argument('--slim_config', default=None, type=str, help='Slim configuration file of slim method')
|
|
parser.add_argument('--opset', type=int, default=16, help='ONNX opset version')
|
|
parser.add_argument('--simplify', action='store_true', help='ONNX simplify model')
|
|
parser.add_argument('--dynamic', action='store_true', help='Dynamic batch-size')
|
|
parser.add_argument('--batch', type=int, default=1, help='Static batch-size')
|
|
args = parser.parse_args()
|
|
if not os.path.isfile(args.weights):
|
|
raise SystemExit('Invalid weights file')
|
|
if args.dynamic and args.batch > 1:
|
|
raise SystemExit('Cannot set dynamic batch-size and static batch-size at same time')
|
|
elif args.dynamic:
|
|
args.batch = None
|
|
return args
|
|
|
|
|
|
if __name__ == '__main__':
|
|
FLAGS = parse_args()
|
|
main(FLAGS)
|