A lot of these I already know, but I thought it would be useful to have a list of not only libraries I need to learn, but their importance in doing so.
I know a lot of these libraries already, but I like writing down references anyway so…
Main Checklist for Reference Notes
Math:
Stats:
Comp Neuro: https://neuralensemble.org/
ChatGPT Generated
Computational Neuroscience
| Library | Why You Need It |
|---|---|
| Brian2 | THE simulator for spiking neural networks — fast, very Pythonic, fantastic for experimenting with neuron models. |
| NEURON + LFPy | If you need detailed, biologically-realistic neuron modeling (ion channels, dendrites, etc.) — used in “heavy” neuroscience papers. |
| NEST | Large-scale spiking neural network simulations. Good if you’re scaling up to networks of thousands/millions of neurons. |
| Elephant | Data analysis for neuroscience (spike trains, time series, STA, etc.). Works well with Neo. |
| Neo | Data structures for electrophysiology data (standardizes how spikes, events, analog signals, etc. are stored). |
| BindsNET | Deep learning + spiking neural network toolkit. If you want SNNs and ML connections. |
| PyNN | Abstraction layer to write one code that can run on Brian2, NEST, NEURON, etc. |
Math / Numerical Methods
| Library | Why You Need It |
|---|---|
| NumPy | Obviously. Vectorized arrays, linear algebra, basic numerics. |
| SciPy | Signal processing, integration, ODE solvers, optimization — especially important for modeling neurons and systems. |
| SymPy | Symbolic math (algebra, calculus, solving equations symbolically). Useful for deriving equations before coding. |
| JAX | If you want autograd (automatic differentiation) and insane speedups with GPU/TPU acceleration for numerical work. |
| mpmath | Arbitrary-precision math. Helps if you need very high precision (rare but sometimes essential). |
| CVXPY | Convex optimization modeling, if you’re solving optimization problems (important for fitting models, control theory, etc.) |
Data and Visualization
| Library | Why You Need It |
|---|---|
| Pandas | Essential for managing experimental data, behavioral recordings, trial logs, spike data, etc. |
| Matplotlib (with mpl_toolkits for 3D) | Standard, but flexible enough if you get good with it (not just basic plots). |
| Seaborn | Statistical data visualization built on top of matplotlib. Very clean for correlation matrices, kernel density plots, etc. |
| Plotly | If you need interactive 2D/3D plots (like dragging around brain activity plots). |
| h5py | Reading and writing .h5 (HDF5) files. Essential when datasets get too big for CSVs and small formats. |
| scikit-learn | Standard ML algorithms and preprocessing — not for big models, but awesome for exploratory data work (e.g., PCA, clustering neural activity). |
| xarray | Multidimensional labeled arrays — great for managing very high-dimensional simulations, like time × space × neurons × trials. |
| PyTables | Advanced handling of very large datasets stored in HDF5 format. |
🔥 Bonus (Power Tools You Might Not Know)
| Library | Why You Need It |
|---|---|
| Better package management than pip/conda for scientific projects. | |
| Numba | Just-in-time (JIT) compiler for speeding up inner loops that can’t be vectorized. |
| dask | Parallelize data operations when your data doesn’t fit in memory (and way easier than pure multiprocessing). |
| NetworkX | Graph theory, essential if you’re modeling brain connectivity as a network (very common). |
| PyMC / ArviZ | Bayesian modeling, probabilistic programming. Crucial for uncertainty quantification. |
| OpenAI Gym / PettingZoo | If you later branch into neural control tasks, simulations of environments are here. |
| Scikit-image | For analysis if you’re working with microscopy or brain imaging data. |
| statsmodels | Advanced statistical tests and models, beyond what scikit-learn usually covers. |
If you only want a “working core” you can expand later:
-
numpy,scipy,matplotlib,pandas -
brian2,neo,elephant -
sympy,jax -
seaborn,h5py,scikit-learn -
dask(for larger data)
General
| Rank | Library | Primary Use Case | Learning Source |
|---|---|---|---|
| 1 | NumPy | Scientific Computing | documentation |
| 2 | Pandas | Data Analysis | documentation |
| 3 | Matplotlib | Data Visualization | documentation |
| 4 | SciPy | Scientific Computing | |
| 5 | Scikit-learn | Machine Learning | |
| 6 | TensorFlow | Machine Learning/AI | |
| 7 | Keras | Machine Learning/AI | |
| 8 | PyTorch | Machine Learning/AI | |
| 9 | Flask | Web Development | |
| 10 | Django | Web Development | |
| 11 | Requests | HTTP for Humans | |
| 12 | BeautifulSoup | Web Scraping | |
| 13 | Selenium | Web Testing/Automation | |
| 14 | PyGame | Game Development | |
| 15 | SymPy | Symbolic Mathematics | |
| 16 | Pillow | Image Processing | |
| 17 | SQLAlchemy | Database Access | |
| 18 | Plotly | Interactive Visualization | |
| 19 | Dash | Web Applications | |
| 20 | Jupyter | Interactive Computing | |
| 21 | FastAPI | Web APIs | |
| 22 | PySpark | Big Data Processing | |
| 23 | NLTK | Natural Language Processing | |
| 24 | spaCy | Natural Language Processing | |
| 25 | Tornado | Web Development | |
| 26 | Streamlit | Data Apps | |
| 27 | Bokeh | Data Visualization | |
| 28 | PyTest | Testing Framework | |
| 29 | Celery | Task Queuing | |
| 30 | Gunicorn | WSGI HTTP Server | |
Neuroscience Tools
- Nengo - Library for creating and simulating large-scale brain models.
- Nitime - Timeseries analysis for neuroscience data.
- Nilearn - Module for performing statistical learning/machine learning on NeuroImaging data.
- DIPY - Toolbox for analysis of MR diffusion imaging.
- MNE-Python - Community-driven software for processing time-resolved neural signals including electroencephalography (EEG) and magnetoencephalography (MEG).
- NiBabel - Provides read and write access to some common medical and neuroimaging file formats.
- PsychoPy - Package for running psychology and neuroscience experiments. It allows for creating psychology stimuli in Python.
- NuPic - Numenta Platform for Intelligent Computing is an implementation of Hierarchical Temporal Memory (HTM), a theory of intelligence based strictly on the neuroscience of the neocortex.
- Brian2 - Free, open source simulator for spiking neural networks.
- expyriment - Platform-independent lightweight Python library for designing and conducting timing-critical behavioural and neuroimaging experiments.
- BindsNET - Package for simulating spiking neural networks for reinforcement & machine learning.
- SpikeInterface - Framework designed to unify spike-sorting technologies
- NiMARE - NiMARE is a Python package for neuroimaging meta-analyses