DNN_tools.py

import os
import sys
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
import time

# 保存图片的一些设置
isShowPic = 1
Leftp = 0.18
Bottomp = 0.18
Widthp = 0.88-Leftp
Heightp = 0.9-Bottomp
pos = [Leftp, Bottomp, Widthp, Heightp]


#  图片保存函数
def mySaveFig(pltm, fntmp, fp=0, ax=0, isax=0, iseps=0, isShowPic=0):
    if isax == 1:
        pltm.rc('xtick', labelsize=18)
        pltm.rc('ytick', labelsize=18)
        ax.set_position(pos, which='both')
    fnm = '%s.png'%(fntmp)
    pltm.savefig(fnm)
    if iseps:
        fnm = '%s.eps'%(fntmp)
        pltm.savefig(fnm, format='eps', dpi=600)
    if fp != 0:
        fp.savefig("%s.pdf"%(fntmp), bbox_inches='tight')
    if isShowPic == 1:
        pltm.show()
    elif isShowPic == -1:
        return
    else:
        pltm.close()


#  日志记数
def log_string(out_str, log_out):
    log_out.write(out_str + '\n')  # 将字符串写到文件log_fileout中去，末尾加换行
    log_out.flush()                # 清空缓存区
    # flush() 方法是用来刷新缓冲区的，即将缓冲区中的数据立刻写入文件，同时清空缓冲区，不需要是被动的等待输出缓冲区写入。
    # 一般情况下，文件关闭后会自动刷新缓冲区，但有时你需要在关闭前刷新它，这时就可以使用 flush() 方法。


def print_and_log_train_one_epoch(i_epoch, run_time, learn_rate, penalty_bd, penalty_powu, pwb, loss_it_tmp, loss_bd_tmp,
                                  loss_tmp, udu, train_mse_tmp, train_res_tmp, log_out=None):
    # 将运行结果打印出来
    print('train epoch: %d, time: %.3f' % (i_epoch, run_time))
    print('learning rate: %.10f' % learn_rate)
    print('boundary penalty: %.10f' % penalty_bd)
    print('penalty for product of normal and scale: %.10f' % penalty_powu)
    print('weights and biases with  penalty: %.10f' % pwb)
    print('loss_it for training: %.10f' % loss_it_tmp)
    print('loss_bd for training: %.10f' % loss_bd_tmp)
    print('loss for training: %.10f' % loss_tmp)
    print('The product of normal and scale for training: %.10f' % udu)
    print('solution mean square error for training: %.10f' % train_mse_tmp)
    print('solution residual error for training: %.10f\n' % train_res_tmp)

    log_string('train epoch: %d,time: %.10f' % (i_epoch, run_time), log_out)
    log_string('learning rate: %.10f' % learn_rate, log_out)
    log_string('boundary penalty: %.10f' % penalty_bd, log_out)
    log_string('penalty for product of normal and scale: %.10f' % penalty_powu, log_out)
    log_string('weights and biases with  penalty: %.10f' % pwb, log_out)
    log_string('loss_it for training: %.10f' % loss_it_tmp, log_out)
    log_string('loss_bd for training: %.10f' % loss_bd_tmp, log_out)
    log_string('loss for training: %.10f' % loss_tmp, log_out)
    log_string('The product of normal and scale for training: %.10f' % udu, log_out)
    log_string('solution mean square error for training: %.10f' % train_mse_tmp, log_out)
    log_string('solution residual error for training: %.10f\n' % train_res_tmp, log_out)


def print_and_log_test_one_epoch(mse2test, res2test, log_out=None):
    # 将运行结果打印出来
    print('mean square error of predict and real for testing: %.10f' % mse2test)
    print('residual error of predict and real for testing: %.10f\n' % res2test)

    log_string('mean square error of predict and real for testing: %.10f' % mse2test, log_out)
    log_string('residual error of predict and real for testing: %.10f\n\n' % res2test, log_out)