Update: bug fixes

AchintyaX · Sep 11, 2022 · 1be88dd · 1be88dd
1 parent 5e3dc63
commit 1be88dd
Show file tree

Hide file tree

Showing 8 changed files with 14 additions and 13 deletions.
diff --git a/README.md b/README.md
@@ -5,6 +5,9 @@ You can download the stock data and perform anomaly detection using a range of a
 
 ## Installation
 For installing the pre-release version use -
+` pip install https://github.com/AchintyaX/fad/releases/download/v0.1.0/fad-0.1.0-py3-none-any.whl`
+
+## Usage
 1. fetch_data - Used for fetching stock data using yahoo finance API and transforming it for anamoly detection : 
 ```
 In [1]: from fad import FinancialAnomalyDetector
@@ -36,9 +39,7 @@ Out[6]:
 
 ```
 
-` pip install https://github.com/AchintyaX/fad/releases/download/v0.1.0/fad-0.1.0-py3-none-any.whl`
-
-## Usage
+2. anamly_detection - Used to find anomalies in the data returning the dataframe along with plotting them in the time series
 
 The package has 2 primary methods - 
 1. 

diff --git a/dist/fad-0.1.0-py3-none-any.whl b/dist/fad-0.1.0-py3-none-any.whl
diff --git a/dist/fad-0.1.0.tar.gz b/dist/fad-0.1.0.tar.gz
diff --git a/dist/fad-0.1.1-py3-none-any.whl b/dist/fad-0.1.1-py3-none-any.whl
diff --git a/dist/fad-0.1.1.tar.gz b/dist/fad-0.1.1.tar.gz
diff --git a/fad/anamoly_utils.py → fad/anomaly_utils.py b/fad/anamoly_utils.py → fad/anomaly_utils.py
@@ -13,21 +13,21 @@
 
 def seasonal_ad_detector(data: pd.DataFrame, print_plot: bool = True) -> Optional[pd.DataFrame]:
     seasonal_ad = SeasonalAD(c=seasonal_c, side=side)
-    anamolies = seasonal_ad.fit_detect(data)
+    anomalies = seasonal_ad.fit_detect(data)
 
     if print_plot == True:
-        print(plot(data, anomaly=anamolies, anomaly_color="orange", anomaly_tag="marker"))
+        print(plot(data, anomaly=anomalies, anomaly_color="orange", anomaly_tag="marker"))
 
-    return anamolies
+    return anomalies
 
 def min_cluster_detector(data: pd.DataFrame, print_plot: bool =True) -> Optional[pd.DataFrame]:
-    min_cluster_ad = MinClusterDetector(KMeans(n_clusters=3))
-    anamolies = min_cluster_ad.fit_detect(data)
+    min_cluster_ad = MinClusterDetector(KMeans(n_clusters=n_clusters))
+    anomalies = min_cluster_ad.fit_detect(data)
 
     if print_plot == True:
-        plot(data, anomaly=anamolies, ts_linewidth=1, ts_markersize=3, anomaly_color='red', anomaly_alpha=0.3, curve_group='all')
+        plot(data, anomaly=anomalies, ts_linewidth=1, ts_markersize=3, anomaly_color='red', anomaly_alpha=0.3, curve_group='all')
 
-    return anamolies
+    return anomalies
 
 def outlier_detector(data: pd.DataFrame, print_plot: bool = True) -> Optional[pd.DataFrame]:
     outlier_ad = OutlierDetector(LocalOutlierFactor(contamination=contamination))

diff --git a/fad/financial_ad.py b/fad/financial_ad.py
@@ -1,7 +1,7 @@
 import pandas as pd
 import yfinance as yf 
 import numpy as np 
-from fad.anamoly_utils import seasonal_ad_detector, min_cluster_detector, outlier_detector, level_shift_detector, volatile_shift_detector
+from fad.anomaly_utils import seasonal_ad_detector, min_cluster_detector, outlier_detector, level_shift_detector, volatile_shift_detector
 from adtk.data import validate_series 
 
 
@@ -30,7 +30,7 @@ def fetch_data(self, company_name: str, start_date: str, end_date: str) -> pd.Da
 
         return data
 
-    def anamoly_detect(self, data: pd.DataFrame, method: str = 'min_cluster_ad', print_plot: bool = True) -> pd.DataFrame:
+    def anomaly_detect(self, data: pd.DataFrame, method: str = 'min_cluster_ad', print_plot: bool = True) -> pd.DataFrame:
         detection_method = self.detector_map[method]
 
         anomalies = detection_method(data=data, print_plot=print_plot)

diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "fad"
-version = "0.1.0"
+version = "0.1.1"
 description = "experimentation with anamoly detection"
 authors = ["Achintya Shankhdhar <achintya.shankhdhar@simpplr.com>"]
 license = "MIT"