Ncnn: run ncnn2mem show err as: find_blob_index_by_name data failed

Created on 10 May 2018 · 2Comments · Source: Tencent/ncnn

Hi,
i use the pretrain caffe mobilefacenet convert to NCNN, that is OK. But when i use ncnn2mem, error as:

./ncnn2mem mobilefacenet.param mobilefacenet.parm.bin mobilefacenet.id.h mobilefacenet.mem.h

find_blob_index_by_name data failed
段错误 (核心已转储)

could you do me a favor, thanks.

PS.
the head of mobilefacenet_deploy.prototxt as:

name:"Caffe_MobileFaceNet"

input: "data"
input_dim: 1
input_dim: 3
input_dim: 112
input_dim: 96

######## CNN Architecture

layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 64
kernel_size: 3
stride: 2
pad: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "conv1/bn"
type: "BatchNorm"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}

Source

frankieGitHub

Most helpful comment

Change
input: "data"
input_dim: 1
input_dim: 3
input_dim: 112
input_dim: 96
To
layer {
name: "data"
type: "Input"
top: "data"
input_param { shape: { dim: 1 dim: 3 dim: 112 dim: 96 } }
}

lvpchen on 10 May 2018

👍3

All 2 comments

Change
input: "data"
input_dim: 1
input_dim: 3
input_dim: 112
input_dim: 96
To
layer {
name: "data"
type: "Input"
top: "data"
input_param { shape: { dim: 1 dim: 3 dim: 112 dim: 96 } }
}