Home

• General
• ML Issues
• DL Techniques
• NLP
• Time series
• Feature Engineering
• Adversarial example
• Book
• Parallel Computing
• Tensorflow
• Keras
• Bot
• Datasets
• Visualization
• General Experience

General

• Build DS portfolio: https://towardsdatascience.com/how-to-build-a-data-science-portfolio-5f566517c79c
• DeepMind: https://deepmind.com/research/publications/
• Tencent AI Lab: https://ai.tencent.com/ailab/paper-list.html
• Google AI blog: https://ai.googleblog.com/
• Kaggle toxic comment classification challenge: https://medium.com/the-artificial-impostor/review-kaggle-toxic-comment-classification-challenge-part-1-934447339309
• http://blog.kaggle.com/2016/07/21/approaching-almost-any-machine-learning-problem-abhishek-thakur/
• https://cacm.acm.org/magazines/2017/8/219605-data-science/fulltext
• http://www.fast.ai/2016/12/08/org-structure/
• List of conferences:
  • http://www.kdnuggets.com/2013/11/top-conferences-data-mining-data-science.html
  • http://www.guide2research.com/topconf/machine-learning
• Back-propagation: http://neuralnetworksanddeeplearning.com/chap2.html
• Set seed random numbers for Keras: https://machinelearningmastery.com/reproducible-results-neural-networks-keras/

ML Issues

• https://www.quora.com/What-is-the-vanishing-gradient-problem
• http://www.kdnuggets.com/2017/07/when-not-use-deep-learning.html
• Human accuracy: http://karpathy.github.io/2014/09/02/what-i-learned-from-competing-against-a-convnet-on-imagenet/
• DEBUG: https://blog.slavv.com/37-reasons-why-your-neural-network-is-not-working-4020854bd607

DL Techniques

• https://github.com/philipperemy/keras-attention-mechanism
• https://jhui.github.io/2017/11/03/Dynamic-Routing-Between-Capsules/
• http://yerevann.com/a-guide-to-deep-learning/
• http://www.kdnuggets.com/2017/06/applying-deep-learning-real-world-problems.html
• https://medium.com/@aidangomez/let-s-do-this-f9b699de31d9
• https://distill.pub/2016/augmented-rnns/

NLP

• http://microposts2016.seas.upenn.edu/challenge.html
• http://www.semantic-web-journal.net/system/files/swj1562.pdf

Time series

• https://codeburst.io/neural-networks-for-algorithmic-trading-volatility-forecasting-and-custom-loss-functions-c030e316ea7e

Feature engineering

• https://www.analyticsvidhya.com/blog/2017/04/feature-engineering-in-iot-age-how-to-deal-with-iot-data-and-create-features-for-machine-learning/
• For CNN: https://medium.com/merantix/picasso-a-free-open-source-visualizer-for-cnns-d8ed3a35cfc5
• For time series: https://www.analyticsvidhya.com/blog/2017/04/feature-engineering-in-iot-age-how-to-deal-with-iot-data-and-create-features-for-machine-learning/

Adversarial example

• https://blog.openai.com/adversarial-example-research/
• https://github.com/tensorflow/cleverhans
• https://github.com/bethgelab/foolbox

Book

• http://appliedpredictivemodeling.com/
• https://machinelearningcoban.com/2017/03/12/convexity/#-gioi-thieu

Parallel Computing

• https://medium.com/@ageitgey/quick-tip-speed-up-your-python-data-processing-scripts-with-process-pools-cf275350163a
• http://matthewrocklin.com/blog/work/2017/10/16/streaming-dataframes-1?utm_campaign=Data%2BElixir&utm_medium=email&utm_source=Data_Elixir_154
• http://dask.pydata.org

Auto-tuning

• https://github.com/maxpumperla/hyperas

Tensorflow

• https://medium.com/@liyin_27935/tensorflow-cpus-and-gpus-configuration-9c223436d4ef
• Tutorials:
  • https://github.com/vahidk/EffectiveTensorflow#basics

Keras

• Theano 2 tensorflow: https://github.com/keras-team/keras/wiki/Converting-convolution-kernels-from-Theano-to-TensorFlow-and-vice-versa

Install GPU

• https://www.tensorflow.org/install/install_linux
• https://stackoverflow.com/questions/42013316/after-building-tensorflow-from-source-seeing-libcudart-so-and-libcudnn-errors

Bot

• https://www.botlibre.com/forum-post?id=5015
• http://www.zachwhalen.net/posts/how-to-make-a-twitter-bot-with-google-spreadsheets-version-04/
• https://botwiki.org/bots/
• https://support.twitter.com/articles/76915
• https://dev.twitter.com/overview/terms/agreement-and-policy

Datasets

• https://www.w3.org/community/sentiment/wiki/Datasets

Visualization

• https://github.com/plotly/dash
• http://cds.nyu.edu/wp-content/uploads/2014/04/bertini_datascience_showcase_May12_2014.pdf
• Tensorboard - Embedding Visualization: http://www.pinchofintelligence.com/simple-introduction-to-tensorboard-embedding-visualisation/
  • Running: tensorboard --logdir logs/o_6.3-3.1.1.zip/ --host 0.0.0.0
• Tensorboard - plugin - for computer vision:
  • https://github.com/chrisranderson/beholder
  • https://research.googleblog.com/2017/09/build-your-own-machine-learning.html?utm_content=buffer32c43&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
• For text: http://ahmedbesbes.com/how-to-mine-newsfeed-data-and-extract-interactive-insights-in-python.html
• For keras model training - changes of weights when training: https://github.com/brannondorsey/keras_weight_animator
• For visualizing data with numerical and categorical features: https://github.com/pair-code/facets
• CNN visualization:
  • https://arxiv.org/abs/1311.2901
  • http://cs231n.github.io/understanding-cnn/

Google cloud

Running jupyter notebook on cloud VM instance

• https://stharrold.github.io/20151208-ipynb-on-gce-from-chrome.html

Connecting to Linux instances

• https://cloud.google.com/compute/docs/instances/connecting-to-instance#standardssh

Download directories/files from google cloud storage

• gsutil -m cp -r gs://[BUCKET_NAME]/PATH/ DESTINATION/

GIT

• https://www.atlassian.com/git/tutorials/undoing-changes
• http://gitready.com/advanced/2009/02/25/keep-either-file-in-merge-conflicts.html
• Merge - accept theirs:
  • git checkout --theirs
  • git commit -a

Imbalance

• Cluster the abundant class: http://www.kdnuggets.com/2017/06/7-techniques-handle-imbalanced-data.html

General Experience

Install python packages

• Find the missing packages
• Using python -c 'import <package>' to check whether the package is installed
• pip install boto==2.47.0 gensim==2.2.0 google-compute-engine==2.4.0
• conda install gensim

Install graphviz

• pip install graphviz pydot-ng
• sudo apt-get install graphviz
• sudo apt-get install -f

Install jupyterlab

• pip install jupyter lab
• jupyter serverextension enable --py jupyterlab --sys-prefix

Choose env on jupyter with conda

• conda install nb_conda

Algorithms for Categorical Data

• https://github.com/Microsoft/LightGBM
• https://github.com/fmfn/BayesianOptimization

Statistics

Relative risk:

• SE(log(RR)) https://stats.stackexchange.com/questions/126727/why-doesnt-standard-error-for-ratios-have-log-in-it

Others

• Process bar: https://github.com/tqdm/tqdm
• Google programming language: https://golang.org/
• https://elitedatascience.com/machine-learning-iteration
• https://huyenchip.com/2017/07/28/confession.html
• Difference between training and practice: https://medium.com/merantix/picasso-a-free-open-source-visualizer-for-cnns-d8ed3a35cfc5