Updated comments.

app.py

@@ -17,7 +17,7 @@ def generate_output(front_img_path, side_img_path):
     front_keypoint_result = predict_pose(front_img_path, "front.jpg")
     side_keypoint_result = predict_pose(side_img_path, "side.jpg")
 
-    # Should we create the image separately? Seems weird to get it as a result from this only to use it in something else below.
+    # TODO: LOW PRIORITY: Should we create the image separately? Seems weird to get it as a result from this only to use it in something else below.
     front_image = front_keypoint_result[0]
     side_image = side_keypoint_result[0]
 
@@ -25,7 +25,10 @@ def generate_output(front_img_path, side_img_path):
     side_keypoint_data = side_keypoint_result[1]
 
     front_seg_mask = calculate_seg_mask(front_img_path)
-    side_rcnn_mask = calculate_seg_mask(side_img_path) # TODO: Is this the correct mask? In the original code there is a function called 'get_rcnn_mask' which is not used anywhere. The name implies that it should be a rcnn mask, but the code actually requests a seg mask.
+    # TODO: Is this the correct mask? In the original code there is a function called 'get_rcnn_mask' which is not used anywhere. 
+    # The name implies that it should be a rcnn mask, but the code actually requests a seg mask.
+    # If we don't need a rcnn mask, then delete the rcnn code from calculate_masks.py and rename side_rcnn_mask to side_seg_mask.
+    side_rcnn_mask = calculate_seg_mask(side_img_path)
 
     measures_data_frame = calculate_all_measures(front_image, side_image, front_keypoint_data, side_keypoint_data, front_seg_mask, side_rcnn_mask)
 

calculate_measures.py

@@ -131,7 +131,8 @@ def calculate_all_measures(front_image, side_image, front_keypoint_data, side_ke
     results_dict['upper_torso_to_body_ratio'] = upper_torso_area / full_side_body_area
     results_dict['upper_torso_to_body_ratio'] = upper_torso_area / full_side_body_area
 
-    # TODO: Temporary to force thigns to work.
+    # TODO: LOW PRIORITY: Not sure if we need this, but it was in the original code and select_body_shape uses it and requires an ID in a lot of places.
+    # Could potentially remove it but will need to update select_body_shape too.
     results_dict['id'] = "1"
 
     results.append(results_dict)

select_body_shape.py

@@ -1,12 +1,10 @@
 import re
-import numpy as np
 import pandas as pd
 from numpy import dot 
 from numpy.linalg import norm
 from body_shape_lookup import body_shape_lookup
 
 BODY_SHAPE_MEASURES = "body_shape_measures_normalised_updated.csv"
-VOLUNTEERS_MEASURES = "volunteers_measures_normalised_updated.csv"
 
 # selecting specific features
 RATIOS_TO_USE = ['shoulder_to_hip_distance',
@@ -34,20 +32,13 @@ def is_match(row):
 def select_body_shape(normalised_body_shape_measures):
     # load the body shape measures
     body_shape_df = pd.read_csv(BODY_SHAPE_MEASURES)
-    # body_shape_df = normalised_body_shape_measures
 
-    # load the volunteers measures
-    # volunteers_df = pd.read_csv(VOLUNTEERS_MEASURES)
+    # load the calculated measures.
     volunteers_df = normalised_body_shape_measures
 
     # select only the columns corresponding to the ratios
     body_shape_ratios = body_shape_df[RATIOS_TO_USE]
 
-    # Create a DataFrame to store the results
-    results_df = pd.DataFrame(columns=["Volunteer_ID", "Rank_1_Body_Shape", "Score_1",
-                                       "Rank_2_Body_Shape", "Score_2",
-                                       "Rank_3_Body_Shape", "Score_3"])
-
     # calculate euclidean distance for each volunteer
     for index, volunteer_row in volunteers_df.iterrows():
         print(f"\nProcessing volunteer {volunteer_row['id']}")