Same issue as following:

https://github.com/una-dinosauria/human-motion-prediction/issues/10

1

NotFoundError (see above for traceback): Failed to create a NewWriteableFile: XX\checkpoint-10.data-00000-of-00001.tempstate3138973305096497355

The problem turned out to be the maximum path length supported by the Windows shell (255 UTF-16 code words), not an NTFS problem.

So I shorten the folder name, then the issue disappears…🙃

After all official installment, error happened:

Could not fetch URL https://pypi.python.org/simple/ssl/: There was a problem confirming the ssl certificate: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:777) - skipping
Could not find a version that satisfies the requirement ssl (from versions: )
No matching distribution found for ssl

I address this problem by adding path to environment 😥

Summary for Hierarchical Attention Network for Document Classification

Using xmind 😆 and some formula by hand

Check out details 👇

Machine Learning Course in Spring 2019

Text Book: Pattern Recognition and Machine Learning, Bishop

Here are my learning notes, to be continued..😝

Introduction

Supervised Learning

• regression: y is continuous
• classification: y is discrete

supervised+X learning

Unsupervised Learning

• Cluster: well-seperated
• Dimensionality reduction

Reinforcement Learning

KL divergence (KL 散度)

good explanation:

• English: http://www.thushv.com/machine-learning/light-on-math-machine-learning-intuitive-guide-to-understanding-kl-divergence/
• Translation: https://zhuanlan.zhihu.com/p/37452654

Question

You are given a binary tree in which each node contains an integer value.

Find the number of paths that sum to a given value.

The path does not need to start or end at the root or a leaf, but it must go downwards (traveling only from parent nodes to child nodes).

The tree has no more than 1,000 nodes and the values are in the range -1,000,000 to 1,000,000.

Example:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

root = [10,5,-3,3,2,null,11,3,-2,null,1], sum = 8
      10
     /  \
    5   -3
   / \    \
  3   2   11
 / \   \
3  -2   1
Return 3. The paths that sum to 8 are:
1.  5 -> 3
2.  5 -> 2 -> 1
3. -3 -> 11

Question

Given a binary tree and a sum, find all root-to-leaf paths where each path’s sum equals the given sum.

Note: A leaf is a node with no children.

Example:

Given the below binary tree and sum = 22,

1
2
3
4
5
6
7

      5
     / \
    4   8
   /   / \
  11  13  4
 /  \    / \
7    2  5   1

Return:

1
2
3
4

[
   [5,4,11,2],
   [5,8,4,5]
]

Machine Learning

• paper
  • [ ] 机器学习与深度学习经典论文整理: https://mp.weixin.qq.com/s/jKK6AwmCMGWgVK0vPX359w
• Reinforcement Learning
  • [ ] UC Berkeley CS294
  • [ ] 强化学习精要

Life

• [ ] Language: English & Korean
• [ ] Dance & Fit

Question

Given a binary tree and a sum, determine if the tree has a root-to-leaf path such that adding up all the values along the path equals the given sum.

Note: A leaf is a node with no children.

Example:

Given the below binary tree and sum = 22,

1
2
3
4
5
6
7

      5
     / \
    4   8
   /   / \
  11  13  4
 /  \      \
7    2      1

return true, as there exist a root-to-leaf path 5->4->11->2 which sum is 22.

[TOC]

Automated Machine Learning

Key words: automatic machine learning, neural architecture search, hyper-parameter optimization, meta-learning, transfer-learning

1. Introduction

Every aspect of machine learning applications, such as feature engineering, model selection, algorithm selection needs to be carefully configured, which is usually involved heavily with human experts.

AutoML attempts to reduce human assistance in the design, selection and implementation of various machine learning tools used in applications’ pipeline.

Examples:

• Auto-sklearn, try a collection of classifiers on a new problem, then get final predictions from an ensemble of them
• Neural architecture search NAS. Success of AlexNet, VGGNet, GoogleNet, ResNet and DenseNet. Can the neural architecture be automatically designed so that good learning performance can be obtained on the given tasks. Reinforcement Learning has been a powerful and promising tool for NAS.
• Automatic feature engineering, aims to construct a new features set, with which the performance of subsequent machine learning tools can be improved. Existing works on this topic include Data Science Machine(DSM), ExploreKit and FeatureHub