Added script to create a user vector from a csv dump of user ratings

WebApp/recs_from_csv.py

@@ -0,0 +1,46 @@
+import argparse
+import csv
+import numpy as np
+from scipy import sparse
+
+
+def load_id_mapper():
+    mapper_id = {}
+    filename = 'static/data/books.csv'
+    with open(filename, "r", encoding='utf8') as f:
+        reader = csv.DictReader(f, delimiter=",")
+        for i, line in enumerate(reader):
+            mapper_id[line['goodreads_book_id']] = int(line['book_id'])
+    return mapper_id
+
+
+def create_user_vector(filename, mapper):
+    user_vec = np.zeros(10000, dtype=np.float)
+    total_valid_reviews = 0
+    total_invalid_reviews = 0
+    with open(filename, 'r', encoding='utf8') as fp:
+        reader = csv.DictReader(fp, delimiter=',')
+        for i, line in enumerate(reader):
+            goodreads_book_id = line['Book Id']
+            if goodreads_book_id in mapper:
+                book_id = mapper[goodreads_book_id]
+                rating = float(line['My Rating'])
+                user_vec[book_id - 1] = rating
+                total_valid_reviews += 1
+            else:
+                print("Couldn't convert {0}".format(line['Title']))
+                total_invalid_reviews += 1
+    print('Converted user with {0} valid and {1} invalid reviews'.format(total_valid_reviews, total_invalid_reviews))
+    return user_vec
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Gets recommendations from an exported library")
+    parser.add_argument('library_file', help="the path to the library file")
+    parser.add_argument('-u', '--username', help="the username to create the file for")
+    args = parser.parse_args()
+
+    mapper_id = load_id_mapper()
+    user_vec = create_user_vector(args.library_file, mapper_id)
+    user_vec = sparse.csr_matrix(user_vec)
+    sparse.save_npz('static/data/cached_users/user_{}.npz'.format(args.username), user_vec)