5,228 stars · MIT (read from /blob/master/LICENSE; plain and unmodified, 'Darío López Padial (Bukosabino)', 2020) · no GitHub releases at all; PyPI 0.11.0 (2023-11-02) against a code date of 2026-03-18 — a release gap, NOT a dead project · Track this in Scout
Adds 43 standard technical-analysis indicators as new columns on a table of prices.
▶Repo detailsthe review · specs · pros & cons · install
What it is
ta is a small Python library that works on a pandas DataFrame, which is the standard way of holding a table of numbers in Python. You give it columns for the open, high, low and close prices plus the volume traded, and it returns the same table with the indicators attached.
What it is good for. Anyone preparing price data — for a chart, for a study, or as the input to something that learns from numbers. It is the easy answer in a field where the fast answer is painful to install, because it is pure Python and needs no compiler and no outside library.
- It installs anywhere in one command, with no compiler and no separate C library to fight. That is the whole reason it exists.
- 43 indicators grouped into volume, volatility, trend, momentum and other, with one function that adds all of them at once for a first look.
- Plain MIT, read from the licence file, with a named copyright holder and nothing added.
- You must supply the exact column names it expects, and fill in any gaps in the data yourself. Give it a table with holes and it returns numbers that look fine and are not.
- ⚠ No release since 2 November 2023, almost three years, while the code moved on 18 March 2026. It publishes no GitHub releases at all, so the Python package index is the only source, and 122 waiting problems and 34 waiting changes have built up behind that version. Fixes that exist in the code are not in the version you install.
- Adding every indicator at once over a long price history creates dozens of new columns of numbers, which can use a surprising amount of memory. The project publishes no figure for this at all. Its package listing also still advertises only Python 3.6 and 3.7, which cannot be right for code written in 2026.
TA-Lib/ta-lib-pythonCovered in Edition 35. The same job with more than 150 indicators and much faster, and the standard choice; the difference is the install, and from 0.6.5 it ships ready-built packages that include the C library.
Track this in Scout
jealous/stockstatsThe same approach of reaching indicators as columns on a table, with more than 50 of them, and actively maintained where ta has gone quiet; needs Python 3.9 or later.
Track this in Scout
nardew/talippThe same set of indicators but recomputed one new price at a time rather than over a whole table, which suits a live feed and is the wrong shape for bulk preparation.
Track this in Scout
pip install --upgrade ta
