Repo directory:

• Projects are split by folders

• PyTorchStuff/nonlinear_regression/nonlinear_regression.md
  • Linear regression to non linear probabilistic neural network.
• PyTorchStuff/pytorch_lightning/pytorch_lightning_regression.md
  • Trying out PyTorch Lightning
• PyTorchStuff/elastic_net/elastic_linear.md
  • Implementing an elastic net model in PyTorch
• FittingMultimodalDistributions/multimodal_beta_pytorch/multimodal_beta.md
  • Fitting a multimodal beta distribution via gradient descent
• FittingMultimodalDistributions/zero_inflated_poisson_pytorch/zero_inflated_poisson.md
  • Fitting a zero-inflated Poisson distribution via gradient descent

• FitDistWithPyro/fit_pyro_distribution.ipynb
  • Simple example fitting a Gaussian distribution to data using Pyro.
• FitDistWithPyro/fit_beta_distribution.ipynb
  • Simple example fitting a beta distribution to data using Pyro.

• TensorflowProbability/nonlinear_regression.ipynb
  • Linear regression to non linear probabilistic neural network.
• TensorflowProbability/fit_gaussian_tfp.ipynb
  • Fitting a normal distribution with tensorflow probability.

• TensorflowStuff/overfitting_nn.ipynb
  • Do Neural Networks overfit?
• FashionCNN/fashion_cnn.ipynb
  • Convolution neural network for predicting the Fashion MNIST dataset.
• FashionCNN/fashion_batch_norm.ipynb
  • Batch normalisation layer applied to the above CNN model

• TimeSeries/neural_prophet/neural_prophet_speed_test.md
  • Speed of fitting and predict of neuralprophet vs fbprophet
• TimeSeries/neural_prophet/arima.md
  • Can we fit long AR models with neuralprophet
• machine_vision/3d_screen
  • Using Google's mediapipe to try simulate a 3D screen
• machine_vision/face_distance
  • Using Google's mediapipe, measure the distance of a face to the screen from a webcam feed

• ParquetDatasets/parquet_datasets.ipynb
  • Exporting dataframes to partitioned parquet files.
• Dask vs multiprocessing
  • Comparing the API fo dask to multiprocessing for general functions
• recommenders/multi_armed_bandits_benchmarks
  • Exploring multi-armed bandit benchmarks
• PCA image compression
  • Using PCA to compress MNIST images
• Dense autoencoder image compression
  • Using a dense autoencoder to compress MNIST images

• BootstrappedRegessionCoefficients/bootstrap_regression.ipynb
  • Confirming theoretical regression coefficient distributions with bootstrapped samples.
• SequentialBayesianRegression/sequential_bayesian_linear_regression.md
  • Sequential Bayesian linear regression model
• SequentialBayesianRegression/adaptive_coefficients.md
  • Bayesian regression adapting to non-stationary data

• CausalInference/causal_regression_coefficient.ipynb
  • Exploring the effect of confounding variables on regression coefficients

• TFL Cycle Analysis
  • Analysis in to the number of bike trips taken per day in London.
• NBA Score Trajectories
  • Flask app to show scores of a basketball match against time.
• Installed energy capacity
  • Analysis into European installed energy capacity
• NBA LeBron Minutes
  • Analysis into LeBron James playing minutes
• NBA Shooting - Kedro project
  • Kedro data pipelines to plot player scoring probability distributions
• NBA Shooting Data
  • Not finished yet - animations of shots taken by Kobe Bryant over time.

The various analysis was built in Python 3.

Some projects have their own requirements/environment. The general setup is installed by:

python3 -m venv dataAnalysisEnv
source dataAnalysisEnv/bin/activate
python -m pip install --upgrade pip
pip install -r requirements.txt

jupyter nbconvert notebook.ipynb --to markdown

This is automated via github actions.

Custom library installed as a dev library for continued development

Use the settings.json file in the repo

Aim is for future work to be incorporated by working on separate branches and merge to master when finished.

• Time series forecasting
  • Greykite
    • https://arxiv.org/abs/2105.01098
    • https://towardsdatascience.com/linkedins-response-to-prophet-silverkite-and-greykite-4fd0131f64cb
    • Imputation of missing regressors
    • Change points in seasonalities
    • Quantiles loss
    • Utilities for diagnosing
    • faster inference
    • Autoregressive
  • LightGBM
  • Orbit
    • https://eng.uber.com/orbit/
• Deep learning
  • Pytorch
  • Embeddings
  • Tensorflow/pytorch - 1D functions
  • FashionMNIST VAE
• Causal inference
• Data validations - great expectations
  • https://github.com/tamsanh/kedro-great
• Computer vision
• NLP
  • Keyword extraction from reviews etc.
  • Sentiment analysis
• Gaussian processes
• Bayesian regression
• Recommender systems
  • Automatic playlist continuation
  • Thompson sampling example
• Switch away from Matplotlib
• Quantile regression in pytorch
  • Lasso regression
  • Dropout better than regularisation?
• GBM/LightGBM
• Data engineering
  • Polars
  • DuckDB
• Docker

• https://archive-beta.ics.uci.edu/ml/datasets#music
• https://ritual.uh.edu/mpst-2018/
  • NLP

• Build project template repo
• Publish interpret-ml piece
• NBA
  • Player position classification model
  • Bayesian sequential team rating
  • Player VAE - how are players related
    • College stats to NBA VAE
• M5/M4 forecasting
  • Walmart demand forecasting
  • with LightGBM
  • Greykite
• PCA via embedding layer
• NN to predict tempo from song, generate dummy dataset
• Word embeddings plot with hiplot
  • Plot with PCA first and compare with hiplot
• Compare linear regression MC dropout to theoretical results
• Optimal car charging schedule
• Media pipe - 3d audio
  • Face distance javascript web app with react
• Covid UK plot against time on a map
  • https://www.reddit.com/r/dataisbeautiful/comments/pay78n/oc_active_covid19_cases_per_capita_in_usa_1212020/
• Autoencoder using transfer learning?
  • what do we use for the decoder?
• Fit a sinusoid to noisy data
  • Fourier
  • Gradient descent
  • MCMC
  • Variational inference
• Double dip loss trajectories
• Fitting NNs to common functions (exp etc.), deep vs wide, number of parameters for given error
• Fit a NN to seasonal data with fourier series components
• DoubleML on heart data to find CATE
• Github action to publish ipynbs to markdown
• Hierarchical models
  • Mixed effects model - is it the same as a fixed effects model (lin/log regression) with one hot encoding for the categorical variables + a fixed effect?
  • Hierarchical bayesian models - for when we have categorical features with share effects over other features
  • Fit with MCMC
  • Similarities to ridge regression - only some coefficients are regularised
  • Generate data and fit each model
  • Ref
    • https://www.youtube.com/watch?v=38yOWMMCeMk&list=WL&index=5
• Bimomial regression = logistic regression
• Linear regression = logistic regression, relationship to Linear Thompson Sampling