Trend Detection in Python. Applicable for real-world industry use cases in time series.

Navigation

Verified details

These details have been verified by PyPI
Project links
GitHub Statistics
Maintainers
Avatar for BarshaDeb from gravatar.com BarshaDeb Avatar for russellsb from gravatar.com russellsb

Unverified details

These details have not been verified by PyPI
Meta
Classifiers
Report project as malware

Project description

PyTrendy Logo

PyTrendy

PyPI version Python License: MIT
Tests Release
codecov Downloads

PyTrendy is a robust solution for identifying and analyzing trends in time series. Unlike other trend detection packages, it is robust to noisy & flat segments, and handles for gradual & abrupt trend cases with a high precision. It aims to be the best package for trend detection in python.

Features

Quickstart

Install the package from PyPi.

pip install pytrendy

Import pytrendy.

import pytrendy as pt

Load daily time series data. In this case, we're using one of pytrendy's custom examples.

df = pt.load_data('series_synthetic')
print(df)

#             date     abrupt    gradual  gradual-noisy-20
#  0    2025-01-01  19.578066  12.500000         27.514106
#  1    2025-01-02  19.358378  13.421717         -6.620099
#  2    2025-01-03  19.228408  13.474026         22.122134
#  3    2025-01-04  19.727130  13.474026         13.863735
#  4    2025-01-05  20.773716  14.505772          8.884535
#  ..          ...        ...        ...               ...
#  176  2025-06-26   4.718725  20.616883         19.790026
#  177  2025-06-27   4.242065  20.978084         19.181404
#  178  2025-06-28   6.012296  22.449495         -6.563936
#  179  2025-06-29   4.603068  23.486652         48.291088
#  180  2025-06-30   4.435105  22.240260          3.343233

Run trend detection & plot the results.

results = pt.detect_trends(df, date_col='date', value_col='gradual', plot=True)

The results object can be used to summarise, further analyse, and generally inspect the trend detections.

results.print_summary()

#  Detected: 
#  - 3 Uptrends. 
#  - 3 Downtrends.
#  - 3 Flats.
#  - 0 Noise.

#  The best detected trend is Down between dates 2025-05-09 - 2025-06-17

#  Full Results:
#  -------------------------------------------------------------------------------
#              direction       start         end  days  total_change  change_rank
#  time_index                                                                   
#  1                 Up  2025-01-02  2025-01-24    22     14.013348            5
#  2               Down  2025-01-25  2025-02-05    11    -13.564214            6
#  3               Flat  2025-02-06  2025-02-09     3           NaN            7
#  4                 Up  2025-02-10  2025-03-14    32     24.632035            3
#  5               Flat  2025-03-15  2025-03-17     2           NaN            8
#  6               Down  2025-03-18  2025-04-01    14    -22.721861            4
#  7                 Up  2025-04-02  2025-05-08    36     72.611833            2
#  8               Down  2025-05-09  2025-06-17    39    -73.253968            1
#  9               Flat  2025-06-18  2025-06-30    12           NaN            9 
#  -------------------------------------------------------------------------------

You can directly call the object as a pandas dataframe. Note change_rank which prioritises long duration and high magnitude of change.

results.df
time_index direction start end trend_class change pct_change days total_change SNR change_rank
1 Up 2025-01-02 2025-01-24 gradual 14.013348 1.044080 22 14.013348 22.207980 5
2 Down 2025-01-25 2025-02-05 gradual -13.564214 -0.554982 11 -13.564214 17.360657 6
3 Flat 2025-02-06 2025-02-09 NaN NaN NaN 3 NaN 20.126008 7
4 Up 2025-02-10 2025-03-14 gradual 26.015512 1.974942 32 24.632035 18.871430 3
5 Flat 2025-03-15 2025-03-17 NaN NaN NaN 2 NaN 17.350339 8
6 Down 2025-03-18 2025-04-01 gradual -22.721861 -0.591909 14 -22.721861 16.762790 4
7 Up 2025-04-02 2025-05-08 gradual 73.687771 3.944243 36 72.611833 21.701162 2
8 Down 2025-05-09 2025-06-17 gradual -73.253968 -0.805442 39 -73.253968 21.122099 1
9 Flat 2025-06-18 2025-06-30 NaN NaN NaN 12 NaN 19.418124 9

Upcoming

  • Full documentation with all features [WIP].
  • Automated testing in CI/CD pipeline with full code coverage.
  • Even more robust edge case testing & generalising.
  • Customising more options for windows.

Project details

Verified details

These details have been verified by PyPI
Project links
GitHub Statistics
Maintainers
Avatar for BarshaDeb from gravatar.com BarshaDeb Avatar for russellsb from gravatar.com russellsb

Unverified details

These details have not been verified by PyPI
Meta
Classifiers

Release history Release notifications | RSS feed

1.2.0

1.2.0.dev1 pre-release

1.1.11

1.1.11.dev4 pre-release

1.1.11.dev3 pre-release

1.1.11.dev2 pre-release

1.1.11.dev1 pre-release

This version

1.1.10

1.1.9

1.1.8

1.1.7

1.1.6

1.1.5

1.1.4

1.1.3

1.1.2

1.1.1

1.0.8

1.0.7

1.0.6

1.0.5

1.0.4

1.0.3

1.0.2

1.0.1

0.0.12

0.0.11

0.0.10

0.0.9

0.0.8

0.0.7

0.0.5

0.0.4

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pytrendy-1.1.10.tar.gz (33.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pytrendy-1.1.10-py3-none-any.whl (41.4 kB view details)

Uploaded Python 3

File details

Details for the file pytrendy-1.1.10.tar.gz.

File metadata

  • Download URL: pytrendy-1.1.10.tar.gz
  • Upload date:
  • Size: 33.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for pytrendy-1.1.10.tar.gz
Algorithm Hash digest
SHA256 a16d221fdf8ce101b90b8cd6301b73d658fe1feb0ddfaae1218ce0635ea68d69
MD5 b4c453ffa8c8816988e50037b5c7dcb4
BLAKE2b-256 bb19069e6cd6b0ffa1084e908b0308d49180947c9f00e20ae97091ab55f105c9

See more details on using hashes here.

Provenance

The following attestation bundles were made for pytrendy-1.1.10.tar.gz:

Publisher: release.yaml on RussellSB/pytrendy

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file pytrendy-1.1.10-py3-none-any.whl.

File metadata

  • Download URL: pytrendy-1.1.10-py3-none-any.whl
  • Upload date:
  • Size: 41.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for pytrendy-1.1.10-py3-none-any.whl
Algorithm Hash digest
SHA256 f5943e44192a370a77c11b4073dbd072d31a5e8640f13686e89aad0b9569b6ea
MD5 9939d23d96256d90bbee8c7264011046
BLAKE2b-256 afe9ea7acc29d4a9cee7e3f2c3d27e696470547ddb18e2e21cff29e01b13caec

See more details on using hashes here.

Provenance

The following attestation bundles were made for pytrendy-1.1.10-py3-none-any.whl:

Publisher: release.yaml on RussellSB/pytrendy

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page