-
Notifications
You must be signed in to change notification settings - Fork 187
Description
onnx模型输入属性:输入4: 输入节点0名称:mix 输入1: 形状:[1,257,1,2] ;输入节点1名称:conv_cache 输入2:形状:[2,3,1,1,16] ;输入节点2名称:tra_cache 输入3形状:[2,3,1,1,16];输入节点3名称:inter_cache 输入4形状[2,1,3,16];
转换为rknn模型后;model input num: 4, output num: 4
input tensors:
index=0, name=mix, n_dims=4, dims=[1, 1, 2, 257], n_elems=514, size=1028, fmt=NHWC, type=FP16, qnt_type=AFFINE, zp=0, scale=1.000000
index=1, name=conv_cache, n_dims=5, dims=[2, 1, 16, 16, 33], n_elems=16896, size=33792, fmt=UNDEFINED, type=FP16, qnt_type=AFFINE, zp=0, scale=1.000000
index=2, name=tra_cache, n_dims=4, dims=[2, 1, 1, 3], n_elems=6, size=12, fmt=UNDEFINED, type=FP16, qnt_type=AFFINE, zp=0, scale=1.000000
index=3, name=inter_cache, n_dims=4, dims=[2, 33, 16, 1], n_elems=1056, size=2112, fmt=NHWC, type=FP16, qnt_type=AFFINE, zp=0, scale=1.000000
output tensors:
。onnx输入维度与rknn输入维度不一样。什么原因导致