wut0n9

[TOC]

• 1学习任务
  • 1.1训练过程
  • 1.2 推理过程
    • 1.2.1推理过程的优化
• 2 任务求解
• 3评价指标
• 3.1困混度

wut0n9

解决zsh终端无匹配错误


# .zshrc
添加
setopt no_nomatch 

source .zshrc

# 查找版本号、channel

wut0n9

https://github.com/ekalinin/github-markdown-toc

./gh-md-toc ~/projects/Dockerfile.vim/README.md   

wut0n9

checkpoint_file = 'topic_dnn/tf/model-10'
session_conf = tf.ConfigProto(
    allow_soft_placement=True,
    log_device_placement=True
)
with tf.Session(config=session_conf) as sess:
    saver = tf.train.import_meta_graph('{}.meta'.format(checkpoint_file))
    saver.restore(sess, checkpoint_file)
    input_x = tf.get_default_graph().get_operation_by_name('input_x').outputs[0]

wut0n9

learning_rate = 0.1
decay_rate = 0.96
global_step = tf.Variable(0, trainable=False)  # 传入优化器实例的minimize方法，系统自1起增1。

# new_learning_rate = learning_rate * decay_rate^(global_step/decay_step)
# 每迭代decay_steps调度学习率
# staircase=True表示结果取整
learning_rate_decay_scheduler = tf.train.exponential_decay(learning_rate=learning_rate, 
                                                           global_step=global_step, 

wut0n9

#!/usr/bin/env bash


KEYWORD="PROC_NAME"
theadPidList=$(ps aux | grep ${KEYWORD} | grep -v grep | awk '{print $2}')

for threadPid in ${theadPidList};
do
    sudo kill -9 ${threadPid}
    echo "Thread ${threadPid} is killed"

wut0n9

def padding_vector(embedding):
    """
    添加OOV默认词向量
    :param embedding: 
    :return: 
    """
    alpha = 0.5 * (2.0 * np.random.random() - 1.0)
    curr_embed = (2.0 * np.random.random_sample([embedding.shape[1]]) - 1.0) * alpha
    return np.row_stack((embedding, curr_embed))

wut0n9

# https://colab.research.google.com/github/Hvass-Labs/TensorFlow-Tutorials/blob/master/04_Save_Restore.ipynb#scrollTo=WTQRVlJU_1NN
# https://github.com/Hvass-Labs/TensorFlow-Tutorials/blob/master/04_Save_Restore.ipynb


%%matplotlibmatplot  inline
import matplotlib.pyplot as plt
import tensorflow as tf
import numpy as np
from sklearn.metrics import confusion_matrix
import time

wut0n9

%matplotlib inline
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix



def print_confusion_matrix(true_cls, pred_cls, cls_name):
    """
    true_cls、true_cls、pred_cls、cls_name都不是标签的id，而是原始标签文本
    true_cls = [u'真实标签1', u'真实标签2', u'真实标签3',..]

wut0n9

def generate_batch(batch_size, data_vec, word_to_int):
    n_chunk = len(data_vec) // batch_size
    x_batches = []
    y_batches = []
    for i in range(n_chunk):
        start_index = i * batch_size
        end_index = start_index + batch_size


        batches = data_vec[start_index:end_index]