Chart Patterns

https://github.com/zeta-zetra/chart_patterns
Admin
Chart Patterns is a Python library for automated detection of well-known chart patterns in financial...
Tokens:5,621
Snippets:36
Trust Score:5.8
Update:7 months ago
Show doc for...
Tokens:
Raw
### Execute Complete Pattern Detection Workflow

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

A full pipeline example that loads OHLC data, runs multiple pattern detection algorithms, and iterates through results to generate summary statistics and visualizations.

```python
import pandas as pd
from chart_patterns.doubles import find_doubles_pattern
from chart_patterns.triangles import find_triangle_pattern
from chart_patterns.pennant import find_pennant

ohlc = pd.read_csv("eurusd-4h.csv")
ohlc = find_doubles_pattern(ohlc, double="both", progress=True)
ohlc = find_triangle_pattern(ohlc, triangle_type="ascending", progress=True)
ohlc = find_pennant(ohlc, progress=True)

for pattern in ['double', 'triangle', 'pennant']:
    count = len(ohlc[ohlc['chart_type'] == pattern])
    if count > 0:
        display_chart_pattern(ohlc, pattern=pattern, save=True)
```

--------------------------------

### Detect and Visualize Chart Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This snippet demonstrates how to load OHLC financial data from a CSV file, apply the pennant pattern detection algorithm, and render the resulting chart. It requires a pandas DataFrame with 'open', 'high', 'low', and 'close' columns.

```python
import pandas as pd

# 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_pennant(ohlc, progress=True)

# Plot the results 
display_chart_pattern(ohlc, pattern="pennant")
```

--------------------------------

### Detect Inverse Head and Shoulders Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This snippet demonstrates the detection of Inverse Head and Shoulders patterns using the `chart_patterns` library in Python. It employs `find_inverse_head_and_shoulders` for pattern identification and `display_chart_pattern` for visualization. The output plots are saved to the 'images/ihs' directory if multiple patterns are detected.

```python
import pandas as pd
from chart_patterns.chart_patterns.inverse_head_and_shoulders 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  
```

--------------------------------

### Detect Double Tops and Bottoms Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This snippet demonstrates how to use the `find_doubles_pattern` function from the `chart_patterns` library to detect double tops and double bottoms in OHLC data. It requires pandas for data manipulation and the `chart_patterns` library for pattern detection and plotting. The function takes OHLC data and a 'double' type ('tops' or 'bottoms') as input and returns the OHLC data with detected patterns. Plotting is handled by `display_chart_pattern`.

```python
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  
```

--------------------------------

### Detect Triangle Patterns (Ascending, Descending, Symmetrical) with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This Python code snippet shows how to detect various triangle patterns (ascending, descending, symmetrical) using the `find_triangle_pattern` function from the `chart_patterns` library. It requires pandas for data handling and `display_chart_pattern` for visualization. Detected patterns are saved in the 'images/triangle' directory.

```python
import pandas as pd
from chart_patterns.chart_patterns.triangles 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  
```

--------------------------------

### Detect and Analyze Pennant Patterns in Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Demonstrates how to identify pennant patterns within a pandas DataFrame using specific geometric constraints. It includes logic to filter detected patterns and extract detailed pivot point information.

```python
ohlc = find_pennant(
    ohlc,
    lookback=20,
    min_points=3,
    r_max=0.9,
    r_min=0.9,
    slope_max=-0.0001,
    slope_min=0.0001,
    lower_ratio_slope=0.95,
    upper_ratio_slope=1.0,
    progress=True
)

pennants = ohlc[ohlc['chart_type'] == 'pennant']
for idx, row in pennants.iterrows():
    print(f"Pennant at index {idx}: Highs={row['pennant_highs']}, Lows={row['pennant_lows']}")
```

--------------------------------

### Detect Flag Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This code snippet shows how to detect flag chart patterns using the `find_flag_pattern` function from the `chart_patterns` Python library. It requires pandas for reading OHLC data and the `chart_patterns` library for pattern detection and visualization. The `display_chart_pattern` function is used to plot the detected flag patterns, saving them to an 'images/flag' directory if multiple patterns are found.

```python
import pandas as pd
from chart_patterns.chart_patterns.flag 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  
```

--------------------------------

### Detect Pennant Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This code snippet demonstrates how to detect pennant chart patterns using the `find_pennant` function from the `chart_patterns` Python library. It requires pandas for reading OHLC data and `display_chart_pattern` for plotting the detected patterns. The function processes OHLC data and can save plots to an 'images/pennant' directory if multiple patterns are found.

```python
import pandas as pd
from chart_patterns.chart_patterns.pennant import find_pennant
from chart_patterns.chart_patterns.plotting import display_chart_pattern


```

--------------------------------

### Detect Double Tops and Bottoms - Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Detects double top and double bottom chart patterns by analyzing sequences of pivot points. Double tops suggest potential bearish reversals, while double bottoms indicate potential bullish reversals. The function requires OHLC data and returns an augmented DataFrame with pattern details.

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

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# Find double bottom patterns
ohlc = find_doubles_pattern(
    ohlc,
    lookback=25,              # Number of periods to analyze
    double="bottoms",         # Options: "tops", "bottoms", "both"
    bottoms_min_ratio=0.98,   # Minimum ratio between trough points
    progress=True             # Show progress bar
)

# Find double top patterns
ohlc = find_doubles_pattern(
    ohlc,
    lookback=25,
    double="tops",
    tops_max_ratio=1.01,      # Maximum ratio between peak points
    progress=True
)

# Check for detected patterns
patterns_found = ohlc[ohlc['chart_type'] == 'double']
print(f"Found {len(patterns_found)} double patterns")

# Access pattern details
for idx, row in patterns_found.iterrows():
    print(f"Pattern at index {idx}: {row['double_type']}")
    print(f"  Pivot indices: {row['double_idx']}")
    print(f"  Pivot values: {row['double_point']}")

# Plot and save all detected patterns to images/double/
display_chart_pattern(ohlc, pattern="double", save=True)
```

--------------------------------

### Detect Head and Shoulders Patterns with Python

Source: https://github.com/zeta-zetra/chart_patterns/blob/main/README.md

This Python code utilizes the `chart_patterns` library to identify Head and Shoulders chart patterns. It imports `find_head_and_shoulders` for detection and `display_chart_pattern` for visualization. The function processes OHLC data, and detected patterns are plotted and saved in the 'images/hs' directory if multiple instances are found.

```python
import pandas as pd
from chart_patterns.chart_patterns.head_and_shoulders import find_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 the head and shoulers pattern
ohlc = find_head_and_shoulders(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  
```

--------------------------------

### Detect Pivot Points in Price Data - Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Identifies pivot highs and lows in OHLC data by comparing candle values with surrounding candles. This is a foundational function for all chart pattern detection. It requires pandas for data manipulation and returns an augmented DataFrame with pivot point indicators.

```python
import pandas as pd
from chart_patterns.pivot_points import find_all_pivot_points
from chart_patterns.plotting import display_pivot_points

# Load OHLC data (columns: open, high, low, close)
ohlc = pd.read_csv("eurusd-4h.csv")

# Find pivot points with default settings (3 candles left/right)
ohlc = find_all_pivot_points(ohlc, left_count=3, right_count=3)

# Access pivot point data
# pivot column values: 0=no pivot, 1=pivot low, 2=pivot high, 3=both
pivot_lows = ohlc[ohlc['pivot'] == 1]
pivot_highs = ohlc[ohlc['pivot'] == 2]

print(f"Found {len(pivot_lows)} pivot lows and {len(pivot_highs)} pivot highs")

# Visualize pivot points on candlestick chart
display_pivot_points(ohlc, plot_obs=500)
```

--------------------------------

### Visualize and Save Chart Patterns with Plotly

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Shows how to render detected patterns on candlestick charts. The library supports saving visualizations as PNG files or displaying them interactively, with support for custom themes.

```python
from chart_patterns.plotting import display_chart_pattern, display_pivot_points

# Save pattern to image
display_chart_pattern(ohlc, pattern="triangle", save=True, lookback=60)

# Custom theme visualization
custom_theme = {"bg_color": "#1a1a2e", "up_color": "#00ff88", "down_color": "#ff4444"}
display_pivot_points(ohlc, theme=custom_theme, plot_obs=300)
```

--------------------------------

### Find Inverse Head and Shoulders Pattern in Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Detects the bullish reversal Inverse Head and Shoulders pattern. It analyzes OHLC data to find three troughs where the middle trough is lower than the surrounding two, connected by a neckline. Requires pandas and specific functions from chart_patterns.

```python
import pandas as pd
from chart_patterns.inverse_head_and_shoulders import find_inverse_head_and_shoulders
from chart_patterns.plotting import display_chart_pattern

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# Find inverse head and shoulders patterns
ohlc = find_inverse_head_and_shoulders(
    ohlc,
    lookback=60,
    pivot_interval=10,
    short_pivot_interval=5,
    head_ratio_before=0.98,       # Head/left shoulder ratio threshold
    head_ratio_after=0.98,        # Head/right shoulder ratio threshold
    upper_slmax=1e-4,             # Maximum neckline slope
    progress=True
)

# Access pattern data
ihs_patterns = ohlc[ohlc['chart_type'] == 'ihs']
print(f"Found {len(ihs_patterns)} inverse head and shoulders patterns")

for idx, row in ihs_patterns.iterrows():
    print(f"Pattern at index {idx}")
    print(f"  Point indices: {row['ihs_idx']}")
    print(f"  Point values: {row['ihs_point']}")
    # Order: left shoulder, left peak, head, right peak, right shoulder

# Plot and save patterns
display_chart_pattern(ohlc, pattern="ihs", save=True, lookback=60)
```

--------------------------------

### Find Flag Pattern in Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Identifies flag continuation patterns in OHLC data. These patterns are characterized by parallel trendlines sloping against the prior trend, typically forming after a strong price move. Requires pandas and specific functions from chart_patterns.

```python
import pandas as pd
from chart_patterns.flag import find_flag_pattern
from chart_patterns.plotting import display_chart_pattern

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# Find flag patterns
ohlc = find_flag_pattern(
    ohlc,
    lookback=25,                  # Candles to analyze
    min_points=3,                 # Minimum pivot points required
    r_max=0.9,                    # R-squared threshold for highs
    r_min=0.9,                    # R-squared threshold for lows
    slope_max=0,                  # Slope threshold for highs
    slope_min=0,                  # Slope threshold for lows
    lower_ratio_slope=0.9,        # Lower slope ratio limit
    upper_ratio_slope=1.05,       # Upper slope ratio limit
    progress=True
)

# Access flag pattern data
flags = ohlc[ohlc['chart_type'] == 'flag']
print(f"Found {len(flags)} flag patterns")

for idx, row in flags.iterrows():
    print(f"Flag at index {idx}")
    print(f"  High pivot indices: {row['flag_highs_idx']}")
    print(f"  Low pivot indices: {row['flag_lows_idx']}")
    print(f"  High slope: {row['flag_slmax']:.6f}")
    print(f"  Low slope: {row['flag_slmin']:.6f}")

# Plot and save flag patterns
display_chart_pattern(ohlc, pattern="flag", save=True)
```

--------------------------------

### Find Head and Shoulders Pattern in Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Detects the bearish reversal Head and Shoulders pattern. It analyzes OHLC data to find three peaks where the middle peak is higher than the surrounding two, connected by a neckline. Requires pandas for data manipulation and specific functions from chart_patterns.

```python
import pandas as pd
from chart_patterns.head_and_shoulders import find_head_and_shoulders
from chart_patterns.plotting import display_chart_pattern

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# Find head and shoulders patterns
ohlc = find_head_and_shoulders(
    ohlc,
    lookback=60,                  # Number of periods to analyze
    pivot_interval=10,            # Candles for pivot detection
    short_pivot_interval=5,       # Secondary pivot interval (must be < pivot_interval)
    head_ratio_before=1.0002,     # Head/left shoulder ratio threshold
    head_ratio_after=1.0002,      # Head/right shoulder ratio threshold
    upper_slmin=1e-4,             # Maximum neckline slope
    progress=True
)

# Access pattern data
hs_patterns = ohlc[ohlc['chart_type'] == 'hs']
print(f"Found {len(hs_patterns)} head and shoulders patterns")

for idx, row in hs_patterns.iterrows():
    print(f"Pattern at index {idx}")
    print(f"  Point indices: {row['hs_idx']}")
    print(f"  Point values: {row['hs_point']}")
    # Order: left shoulder, left trough, head, right trough, right shoulder

# Plot and save patterns
display_chart_pattern(ohlc, pattern="hs", save=True, lookback=60)
```

--------------------------------

### Find Pennant Pattern in Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Identifies pennant continuation patterns in OHLC data. Similar to symmetrical triangles, pennants form after a strong price move and have converging trendlines that meet at an apex. Requires pandas and specific functions from chart_patterns.

```python
import pandas as pd
from chart_patterns.pennant import find_pennant
from chart_patterns.plotting import display_chart_pattern

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# The find_pennant function would be called here, similar to other pattern detection functions.
# Example placeholder for the function call:
# ohlc = find_pennant(ohlc, ...)

# Access pennant pattern data would follow, e.g.:
# pennants = ohlc[ohlc['chart_type'] == 'pennant']
# print(f"Found {len(pennants)} pennant patterns")

# Plot and save pennant patterns would also be called:
# display_chart_pattern(ohlc, pattern="pennant", save=True)
```

--------------------------------

### Detect Triangle Patterns - Python

Source: https://context7.com/zeta-zetra/chart_patterns/llms.txt

Identifies ascending, descending, and symmetrical triangle chart patterns using linear regression on pivot points. These patterns have specific characteristics regarding resistance and support trendlines. The function requires OHLC data and returns an augmented DataFrame with triangle pattern details.

```python
import pandas as pd
from chart_patterns.triangles import find_triangle_pattern
from chart_patterns.plotting import display_chart_pattern

# Load OHLC data
ohlc = pd.read_csv("eurusd-4h.csv")

# Find ascending triangle (flat top, rising bottom)
ohlc = find_triangle_pattern(
    ohlc,
    lookback=25,                  # Candles to look back
    min_points=3,                 # Minimum pivot points required
    rlimit=0.9,                   # R-squared fit threshold
    slmax_limit=0.00001,          # Slope limit for highs
    slmin_limit=0.00001,          # Slope limit for lows
    triangle_type="ascending",    # Options: "ascending", "descending", "symmetrical"
    progress=True
)

# Find descending triangle (falling top, flat bottom)
ohlc = find_triangle_pattern(ohlc, triangle_type="descending", progress=True)

# Find symmetrical triangle (converging lines)
ohlc = find_triangle_pattern(ohlc, triangle_type="symmetrical", progress=True)

# Access triangle pattern data
triangles = ohlc[ohlc['chart_type'] == 'triangle']
for idx, row in triangles.iterrows():
    print(f"Triangle at {idx}: {row['triangle_type']}")
    print(f"  High slope: {row['triangle_slmax']:.6f}")
    print(f"  Low slope: {row['triangle_slmin']:.6f}")

# Plot and save triangle patterns
display_chart_pattern(ohlc, pattern="triangle", save=True)
```
Tokens:
Raw