i3wrappers

greatabel · greatabel · commit 176b8a76be3d · 2018-06-24T22:09:47.000+08:00
diff --git a/Modular-Programming-with-Python/ch5Working with Module Patterns/i3wrappers/detect_unusual_transfers.py b/Modular-Programming-with-Python/ch5Working with Module Patterns/i3wrappers/detect_unusual_transfers.py
@@ -0,0 +1,86 @@
+# detect_unusual_transfers.py
+
+import random
+import numpy_wrapper as npw
+
+#############################################################################
+
+BANK_CODES = ["AMERUS33", "CERYUS33", "EQTYUS44",
+              "LOYDUS33", "SYNEUS44", "WFBIUS6S"]
+
+BRANCH_IDS = ["125000249", "125000252", "125000371",
+              "125000402", "125000596", "125001067"]
+
+#############################################################################
+
+def main():
+    """ Our main program.
+    """
+    # Create 10,000 random transfers.
+
+    days = [1, 2, 3, 4, 5, 6, 7, 8]
+    transfers = [] # List of (day, bank_code, branch_id, amount) tuples.
+
+    for i in range(10000):
+        day       = random.choice(days)
+        bank_code = random.choice(BANK_CODES)
+        branch_id = random.choice(BRANCH_IDS)
+        amount    = random.randint(1000, 1000000)
+
+        transfers.append((day, bank_code, branch_id, amount))
+
+    # Now process the transfers, grouping them by day and building a NumPy
+    # array mapping each branch ID and bank code combination to the total for
+    # that branch and bank for that day.
+
+    transfers_by_day = {}
+    for day in days:
+        transfers_by_day[day] = npw.new(num_rows=len(BANK_CODES),
+                                        num_cols=len(BRANCH_IDS))
+
+    for day,bank_code,branch_id,amount in transfers:
+        array = transfers_by_day[day]
+        row = BRANCH_IDS.index(branch_id)
+        col = BANK_CODES.index(bank_code)
+        array[row][col] = array[row][col] + amount
+
+    # Get the most recent day.
+
+    latest_day = max(days)
+
+    # Collect the transfers for all days other than the latest one.
+
+    transfers_to_average = []
+    for day in days:
+        if day != latest_day:
+            transfers_to_average.append(transfers_by_day[day])
+
+    # Get the transfers for the current day.
+
+    current = transfers_by_day[latest_day]
+
+    # Calculate the average for each day other than the last one.
+
+    average = npw.average(transfers_to_average)
+
+    # Find the entries in the current day which are more than 150% of the
+    # average.
+
+    unusual_transfers = current > average * 1.5
+
+    for row,col in npw.get_indices(unusual_transfers):
+        branch_id   = BRANCH_IDS[row]
+        bank_code   = BANK_CODES[col]
+        average_amt = int(average[row][col])
+        current_amt = current[row][col]
+
+        print("Branch {} transferred ${:,d}".format(branch_id,
+                                                    current_amt) +
+              " to bank {}, average = ${:,d}".format(bank_code,
+                                                     average_amt))
+
+#############################################################################
+
+if __name__ == "__main__":
+    main()
+
diff --git a/Modular-Programming-with-Python/ch5Working with Module Patterns/i3wrappers/numpy_wrapper.py b/Modular-Programming-with-Python/ch5Working with Module Patterns/i3wrappers/numpy_wrapper.py
@@ -0,0 +1,10 @@
+import numpy
+
+def new(num_rows, num_cols):
+    return numpy.zeros((num_rows, num_cols), dtype=numpy.int32)
+
+def average(arrays_to_average):
+    return numpy.mean(numpy.array(arrays_to_average), axis=0)
+
+def get_indices(array):
+    return numpy.transpose(array.nonzero())
