.The trading robot will automatically and continuously issue limit orders to profit from the price difference;
import argparse
import os.path as osp
import sys
sys.path.insert(0,'.')
import torch
from lib.models import model_factory
from configs import set_cfg_from_file
torch.set_grad_enabled(False)
parse=argparse.ArgumentParser()
parse.add_argument('--config',dest='config',type=str,
default='G:/6666Ground_segmentation0813/configs/bisenetv2_city.py',)
parse.add_argument('--weight-path',dest='weight_pth',type=str,
default='G:/6666Ground_segmentation0813/v4_model_final.pth')#最后的pytorch模型
parse.add_argument('--outpath',dest='out_pth',type=str,
default='G:/6666Ground_segmentation0813/model0124.onnx')#转成onnx的路径
args=parse.parse_args()
cfg=set_cfg_from_file(args.config)
if cfg.use_sync_bn:cfg.use_sync_bn=False
net=model_factorycfg.model_type
net.load_state_dict(torch.load(args.weight_pth),strict=False)
net.eval()
#dummy_input=torch.randn(1,3,*cfg.crop_size)
#dummy_input=torch.randn(1,3,1024,2048)
dummy_input=torch.randn(1,3,480,640)#图像的输入尺寸
input_names=['input_image']
output_names=['preds',]
torch.onnx.export(net,dummy_input,args.out_pth,
input_names=input_names,output_names=output_names,
verbose=False,opset_version=11)