ChartPatterns - Automate the detection of chart patterns

Make it easy to detect well-known chart patterns such as ascending triangles, head and shoulders, flag, etc with the chart_patterns python library. Like Thomas Bulkowski says in his book, Encyclopedia of Chart Patterns, "To knowledgeable investors, chart patterns are not squiggles on a price chart; they are the footprints of the smart money." This python package hopeful helps in this regard.

Table of Contents

• Available Chart Patterns
• Installation
• Getting Started
  • Doubles
  • Flag
  • Head and Shoulders
  • Inverse Head and Shoulders
  • Triangles
• Resources

Available Chart Patterns

Here is a list of currently available and soon to be available chart patterns:

• Doubles

• Flag

• Head and Shoulders

• Inverse Head and Shoulders

• Triangles

• Pennant

• Rounding bottom and top

• Triples

• Wedge

Installation

Install the chart_patterns package by cloning this repo. Place the folder in your working directory.

git clone https://github.com/zeta-zetra/chart_patterns

Make sure to install the required packages using the requirements.txt file.

pip install -r requirements.txt

Getting Started

Once you have installed the package, then you can get started. We provide detailed examples of each of the available chart patterns in the package.

Doubles

 import pandas as pd
 from chart_patterns.chart_patterns.doubles import find_doubles_pattern
 from chart_patterns.chart_patterns.plotting import display_chart_pattern



 # read in your ohlc data 
 ohlc = pd.read_csv("eurusd-4h.csv")  #headers include - open, high, low, close

 # Find the double bottom pattern
 ohlc = find_doubles_pattern(ohlc, double="bottoms")

 # Find the double tops pattern
 ohlc = find_doubles_pattern(ohlc, double="tops")
 

 # Plot the results 
 display_chart_pattern(ohlc, pattern="double") # If multiple patterns were found, then plots will saved inside a folder named images/double  

In the find_doubles_pattern function one can change the following:

• The maximum ratio between the peak points in the tops chart pattern. See the tops_max_ratio parameter.
• The minimum ratio between the trough points in the bottoms chart pattern. See the bottoms_min_ratio parameter.

Flag

 import pandas as pd
 from chart_patterns.chart_patterns.doubles import find_flag_pattern
 from chart_patterns.chart_patterns.plotting import display_chart_pattern

 # read in your ohlc data 
 ohlc = pd.read_csv("eurusd-4h.csv")  #headers include - open, high, low, close

 # Plot the results 
 display_chart_pattern(ohlc, pattern="flag") # If multiple patterns were found, then plots will saved inside a folder named images/flag  
 

Head and Shoulders

 import pandas as pd
 from chart_patterns.chart_patterns.doubles import find_head_and_shoulders_pattern
 from chart_patterns.chart_patterns.plotting import display_chart_pattern

 # read in your ohlc data 
 ohlc = pd.read_csv("eurusd-4h.csv")  # headers must include - open, high, low, close

 # Find the head and shoulers pattern
 ohlc = find_head_and_shoulders_pattern(ohlc)

 # Plot the results 
 display_chart_pattern(ohlc, pattern="hs") # If multiple patterns were found, then plots will saved inside a folder named images/hs  

Inverse Head and Shoulders

 import pandas as pd
 from chart_patterns.chart_patterns.doubles import find_inverse_head_and_shoulders
 from chart_patterns.chart_patterns.plotting import display_chart_pattern

 # read in your ohlc data 
 ohlc = pd.read_csv("eurusd-4h.csv")  # headers must include - open, high, low, close

 # Find inversr head and shoulders
 ohlc = find_inverse_head_and_shoulders(ohlc)

 # Plot the results 
 display_chart_pattern(ohlc, pattern="ihs") # If multiple patterns were found, then plots will saved inside a folder named images/ihs  

Triangles

 import pandas as pd
 from chart_patterns.chart_patterns.doubles import find_triangle_pattern
 from chart_patterns.chart_patterns.plotting import display_chart_pattern

 # read in your ohlc data 
 ohlc = pd.read_csv("eurusd-4h.csv")  # headers must include - open, high, low, close

 # Find the ascending triangle
 ohlc = find_triangle_pattern(ohlc)

 # Find the descending triangle
 ohlc = find_triangle_pattern(ohlc, triangle_type="descending")

 # Find the symmetrical triangle 
 ohlc = find_triangle_pattern(ohlc, triangle_type="symmetrical")

 # Plot the results 
 display_chart_pattern(ohlc, pattern="triangle") # If multiple patterns were found, then plots will saved inside a folder named images/triangle  

Resources

We have a YouTube channel where we go through the code of the chart patterns. In addition, we have a get repo with extra code covering other trading related material.