•
Prophet 1.4.0 (PyPI ~15 Aug 2026) entered maintenance mode: bug fixes, dependency bumps, R parity only — no new features. Keep it for classic seasonal business series; for new work, skim alternatives without mashups. Not investment advice.
Prophet 1.4.0 is on PyPI (upload around 15 August 2026). The GitHub README now states clearly: as of v1.4.0, Prophet is in maintenance mode — only bug fixes, dependency bumps, and R package parity with Python will be accepted; no new features are planned. This Hub note is for after-work Python / data / ML learners who already use Prophet on classic seasonal business series. Not investment advice.
The only course CTA is Introduction to Python Programming.
| Still welcome | Not planned |
|---|---|
| Bug fixes | New forecasting features |
| Dependency bumps for a working install | New model families / APIs |
| R changes to stay in parity with Python | A long feature roadmap |
You do not need to rip Prophet out of a notebook that already forecasts weekly retail / holiday-shaped series well. You do need a calm plan before starting a greenfield forecasting stack in 2026+.
pip install "prophet==1.4.0"
python -c "import prophet; print(prophet.__version__)"
Tiny educational sketch (synthetic seasonality only — not a trading signal):
import pandas as pd
from prophet import Prophet
# Synthetic weekly series — educational only. Not investment advice.
ds = pd.date_range("2023-01-01", periods=104, freq="W")
y = [10 + (i % 52) * 0.1 + (i % 7) * 0.05 for i in range(104)]
df = pd.DataFrame({"ds": ds, "y": y})
m = Prophet()
m.fit(df)
future = m.make_future_dataframe(periods=8, freq="W")
print(m.predict(future)[["ds", "yhat"]].tail())
If you are starting new forecasting work, treat Prophet as one mature option among many. You may briefly evaluate other ecosystems (classical statsmodels/seasonal models, modern foundation-model forecasters, or unified forecasting APIs) — without turning this post into a mashup of our separate Chronos / TimesFM / sktime Hub notes. Open those articles on their own if you need depth.
Decision heuristic for tonight: keep Prophet for an existing seasonal business series; pause and survey before betting a new production pipeline on an unmaintained feature surface.
Pin prophet==1.4.0 on one seasonal notebook, or if Python basics are the gap, Introduction to Python Programming.