Merged visualize.py and added geopandas functions

Multiplicative Algorithm/Visualizations.py

@@ -2,6 +2,8 @@ import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
 import config
+import geopandas as gpd
+from shapely.geometry import LineString
 
 
 days_of_week = ['Saturday', 'Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday','Friday']
@@ -41,4 +43,90 @@ def plot_all_signatures_particular_weekday(W, weekday, mean = False):
         imagename = '../Figures/' + str(signature).zfill(2) + days_of_week[weekday][0:3] + '.png'
         plt.savefig(imagename, dpi=300, bbox_inches='tight')  # dpi ideal for viewing on laptop fullscreen
         plt.clf()
+        
+
+def Heatmap(W1, W2):
+    assert W1.shape == W2.shape
+    Ro = np.corrcoef(W1.T,W2.T)[len(W1.T):,:len(W2.T)]
+    plt.imshow(Ro, cmap='hot', interpolation='nearest')
+    plt.colorbar()
+    plt.title('Heatmap for correlation coefficients comparing two runs of W')
+    plt.xlabel('Columns of W1')
+    plt.ylabel('Columns of W2')
+    plt.savefig('Heatmap')
+    plt.show()
+    
+    return None    
+
+
+# Requires full_link_ids.txt, links.csv, and H_trips.txt in running folder
+# Requires Signature_Maps folder within running folder
+def map_signature(signature):
+    filenames = config.generate_filenames('')
+    
+    full_link_ids = np.loadtxt('full_link_ids.txt', dtype=int)
+    H_loaded = gpd.pd.read_csv(filenames['H_trips'], delimiter=' ', names=full_link_ids)
+    link_df = gpd.pd.read_csv(filenames['links'])
+    
+    crs = {'init': 'epsg:4326'}
+    
+    signature_list = H_loaded.loc[signature,:][H_loaded.loc[signature,:]>0].keys().tolist()
+    
+    if len(signature_list)>0:
+        df = link_df[link_df['link_id'].isin(signature_list)]
+        df['geom'] = df.apply(lambda x: LineString([(x.startX,x.startY),(x.endX,x.endY)]), axis=1)
+        geo_df = gpd.GeoDataFrame(df, crs=crs, geometry='geom')
+        location =  'Signature_Maps\sig' + format(signature, '02d')
+        geo_df.to_file(location)
+    else:
+        print('Signature'+str(signature)+'has no links')
+       
+    return None
+
+
+def map_all_signatures():
+    filenames = config.generate_filenames('')
+    
+    full_link_ids = np.loadtxt('full_link_ids.txt', dtype=int)
+    H_loaded = gpd.pd.read_csv(filenames['H_trips'], delimiter=' ', names=full_link_ids)
+    link_df = gpd.pd.read_csv(filenames['links'])
+    
+    crs = {'init': 'epsg:4326'}
+    
+    signature_link_lists = list()
+    link_df = gpd.pd.read_csv(filenames['links'])
+
+    for signature in range(config.RANK):
+        signature_list = H_loaded.loc[signature,:][H_loaded.loc[signature,:]>0].keys().tolist()
+        signature_link_lists.append(signature_list)
+        if len(signature_list)>0:
+            df = link_df[link_df['link_id'].isin(signature_link_lists[signature])]
+            df['geom'] = df.apply(lambda x: LineString([(x.startX,x.startY),(x.endX,x.endY)]), axis=1)
+            geo_df = gpd.GeoDataFrame(df, crs=crs, geometry='geom')
+            location =  'Signature_Maps\sig' + format(signature, '02d')
+            geo_df.to_file(location)
+        else:
+            print('Signature'+str(signature)+'has no links')
+       
+    return None
+
+
+def map_links_with_many_signatures(threshold=7): # threshold for number of signatures for a link considered high
+    filenames = config.generate_filenames('')
+    
+    full_link_ids = np.loadtxt('full_link_ids.txt', dtype=int)
+    H_loaded = gpd.pd.read_csv(filenames['H_trips'], delimiter=' ', names=full_link_ids)
+    link_df = gpd.pd.read_csv(filenames['links'])
+    
+    crs = {'init': 'epsg:4326'}
+    
+    ds = np.sum(H_loaded>0, axis=0)
+    link_ids_with_high_signature_count = ds[ds>=threshold].keys().tolist()
+    
+    df = link_df[link_df['link_id'].isin(link_ids_with_high_signature_count)]
+    df['geom'] = df.apply(lambda x: LineString([(x.startX,x.startY),(x.endX,x.endY)]), axis=1)
+    geo_df = gpd.GeoDataFrame(df, crs=crs, geometry='geom')
+    geo_df.to_file('High_Signature_Map')
+    
+    return None
         
\ No newline at end of file