Home » Content » Data Science » Page 2
Category:

Data Science

ContentData ScienceFeaturedMachine LearningResearch

Bees, lasers, and machine learning

by Jack Simpson
written by Jack Simpson

A couple of years ago I started my PhD at the Australian National University working to quantify honeybee behaviour. We wanted to build a system that could automatically track and compare different groups of bees within the hive.

I took the project as I had a background in biology, beekeeping and programming, and I wanted to work in a lab where I could learn from a supervisor who was incredibly knowledgeable about both biology and software development.

Continue Reading
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine Learning

Notes from ‘A Few Useful Things to Know about Machine Learning’

by Jack Simpson
written by Jack Simpson

I was reading a paper by Pedro Domingos this evening which had some tips and advice for people using machine learning. I’ve written down some bullet points for my own reference and I hope someone else finds it useful. I know I’ve made some of the mistakes he gives advice about avoiding.

Continue Reading
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine Learning

Deep Learning PyData Talk

by Jack Simpson
written by Jack Simpson

Deep learning is a type of machine learning based on neural networks which were inspired by neurons in the brain. The difference between a deep neural network and a normal natural network is the number of ‘hidden layers’ between the input and output layers.

I recently watched an excellent presentation on Deep Learning by Roelof Pieters titled ‘Python for image and text understanding; One model to rule them all!‘ I can recommend watching it, and I’ve written this post for me to put down a few of my own bullet points from the talk for future reference.

Continue Reading
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine Learning

Improving Model Accuracy

by Jack Simpson
written by Jack Simpson

I wrote a few quick bullet points down from the article “8 Proven Ways for improving the “Accuracy” of a Machine Learning Model” for future reference.

Improving Accuracy

  • Add more data
  • Fix missing values
    • Continuous: impute with median/mean/mode
    • Categorical: treat as separate class
    • Predict missing classes with k-nearest neighbours
  • Outliers
    • Delete
    • Bin
    • Impute
    • Treat as separate to the others
  • Feature engineering
    • Transform and normalise: scale between 0-1
    • Eliminate skewness (e.g. log) for algorithms that require normal distribution
    • Create features: Date of transactions might not be useful but day of the week may be
  • Feature selection
    • Best features to use: identify via visualisation or through domain knowledge
    • Significance: Use p-values and other metrics to identify the right values. Can also use dimensionally reduction while preserving relationships in the data
  • Test multiple machine learning algorithms and tune their parameters
  • Ensemble methods: combine multiple week predictors (bagging and boosting)
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine LearningTips & Tutorials

Working with Imbalanced Classes

by Jack Simpson
written by Jack Simpson

I wrote a few quick bullet points down from the article “8 Tactics to Combat Imbalanced Classes in Your Machine Learning Dataset” for future reference.

Tactics

  • Imbalanced datasets occur when you have a class that occurs much more infrequently than the others.
  • If a model ignores the class, it can still achieve a high classification accuracy, but that’s not the result we want
  • Make sure you use a confusion matrix to ensure that you’re getting acceptable accuracy for all your classes
  • One potential solution to this problem is to collect more data
  • Resampling your dataset (can be random or non-random – stratified):
    • Add copies of underrepresented classes (oversampling/sampling with replacement). Useful if you don’t have much data – 10s of thousands or less.
    • Delete instances of classes that occur frequently (undersampling). Handy to use if you have a lot of data – 10s-100s of thousands of instances
  • Try different algorithms: decision trees can perform well on imbalanced datasets
  • Penalised models: Extra penalties for misclassifying minority class. Examples of these algorithms could include penalized-SVM and penalized-LDA.
  • There are areas of research dedicated to imbalanced datasets: can look into anomaly detection and change detection.
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine LearningTips & Tutorials

Assessing machine learning algorithm performance

by Jack Simpson
written by Jack Simpson

I wrote a few quick bullet points down from the article “How To Implement Machine Learning Algorithm Performance Metrics From Scratch With Python“.

Metrics

  • Classification accuracy
    • Test how well predictions of a model do overall
    • accuracy = correct predictions / total predictions
  • Confusion matrix
    • Use to identify how well your predictions did with different classes
    • Very useful if you have an imbalanced dataset
    • I wrote an extremely hacked together confusion matrix for my tag identification software. I had 4 classes (U, C, R, Q) and the confusion matrix shows you what your model predicted against what the real category was.

U

C

R

Q

U

175

17

67

1

C

11

335

14

0

R

26

8

298

0

Q

6

0

3

93
  • Mean absolute error for regression
    • Positive values – the average of how much your predicted value differ from the real value
  • Root mean squared error for regression
    • Square root of the mean of squared differences between the actual and predicted value
    • Squaring the values gives you positive numbers and finding the root lets you compare the values to the original units.
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine LearningTips & Tutorials

Machine Learning Recipes

by Jack Simpson
written by Jack Simpson

I found an excellent tutorial series on Machine Learning on the Google Developers YouTube channel this weekend. It uses Python, scikit-learn and tensorflow and covers decision trees and k-nearest neighbours (KNN).

I really liked the focus on understanding what was going on underneath the hood. I followed along and implemented KNN from scratch and expanded on the base class they described to include the ability to include k as a variable. You can find my implementation in a Jupyter Notebook here.

Sidenote: If you want to visualise the decision tree, you’ll need to install the following libraries. I used homebrew to install graphviz but you could also use a package manger on Linux:


brew install graphviz
pip3 install pydotplus
0 comment
0 FacebookTwitterPinterestEmail
ContentData ScienceMachine Learning

Visual Diagnostics for More Informed Machine Learning

by Jack Simpson
written by Jack Simpson

I recently watched Rebecca Bilbro’s presentation at PyCon 2016 and thought I’d share a few of my short notes from her interesting presentation.

Model Selection Triple

When selecting a model, rather than going with your default favourite method, take 3 things into account:

  • Feature analysis: intelligent feature selection and engineering
  • Model selection: model that makes most sense for problem/domain space
  • Hyperparameter Tuning: once model and features have been selected, select the parameters that result in optimal performance.

Visual Feature Analysis

  • Boxplots are a useful starting tool for looking at all features as they show you:
    • Central tendency
    • Distribution
    • Outliers
  • Histograms let you examine the distribution of a feature
  • Sploms: Pairwise plots of features to identify:
    • pairwise linear, quadratic and exponential relationships between variables
    • Homo/heteroscedasticity
    • How features are distributed relative to each other
  • Raduiz: Plot features around a circle and show how much pull they have
  • Parallel coordinates: lets you visualise multiple variables as line segments – you want to find separating chords which can help with classification

Evaluation Tools

  • Classification heat maps: show you areas where model is performing best
  • ROC-AUC and Prediction Error Plots: Show you which models are performing better
  • Residual plots: Show you which models are doing best and why
  • Gridsearch and validation curves: shows you the performance of a model along the parameters. You can create a visual heatmap for grid search
0 comment
0 FacebookTwitterPinterestEmail
Newer Posts
Older Posts