Skip to content

Instantly share code, notes, and snippets.

View northeastsquare's full-sized avatar
🎯
Focusing

bytemaster northeastsquare

🎯
Focusing
View GitHub Profile
@northeastsquare
northeastsquare / mobilenet_yolov3_lite_train.prototxt
Created September 11, 2020 02:30
MobileNet-YOLO/models/yolov3/mobilenet_yolov3_lite_train.prototxt
name: "MobileNet-YOLO"
layer {
name: "data"
type: "AnnotatedData"
top: "data"
top: "label"
include {
phase: TRAIN
}
@northeastsquare
northeastsquare / CNN_convert_bin_and_print_featuremap.py
Last active June 30, 2021 06:04
nnie 均值、scale, 打印各层特征。使用image_list;使用cfg文件,与nnie生成wk文件的cfg一样,不需要修改。https://github.com/hanson-young/nniefacelib, https://zhuanlan.zhihu.com/p/107548509
#from __future__ import print_function
import sys
sys.path.insert(0, 'D:\\hisi\\30\\HiSVP_PC_V1.1.3.0\\tools\\nnie\\windows\\ruyi_env_setup-2.0.38\\python35\\Lib\\site-packages\\caffe\\python')
import caffe
import pickle
from datetime import datetime
import numpy as np
import struct
import sys, getopt
import cv2, os, re
@northeastsquare
northeastsquare / mobilenet_v2.prototxt
Created March 1, 2019 08:53
mobilenet_v2.prototxt discrip
name: "MobileNet-v2"
input: "data"
input_dim: 1
input_dim: 3
input_dim: 224
input_dim: 224
layer {
name: "conv_1_1"
type: "Convolution"
bottom: "data"
@northeastsquare
northeastsquare / 三级汉字1605.txt
Created November 19, 2018 08:31
三级汉字1605个
𠙶
@northeastsquare
northeastsquare / 二级汉字3000.txt
Last active November 19, 2018 08:33
二级汉字3000个
廿
@northeastsquare
northeastsquare / README.md
Last active November 19, 2018 08:39
通用规范汉字表 (2013)
@northeastsquare
northeastsquare / shufflenetSimple2
Last active May 10, 2018 03:13
shufflenetSimple2
name: "shufflenet"
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
data_param {
name: "shufflenet"
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
data_param {
name: "ResNet-18"
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
name: "shufflenet"
# transform_param {
# scale: 0.017
# mirror: false
# crop_size: 224
# mean_value: [103.94,116.78,123.68]
# }
input: "data"
input_shape {
dim: 1