feat: add progress bar and comment

README.md

@@ -2,7 +2,17 @@
 This Repository contains quantum circuit simutlation using a Matrix-Product-State representation of the quantum state. It is an approximation algorithm to scale linear with the number of qubits and the depth of the circuit. It has been written for the Quantum Computing Seminar at the Technical University of Munich. The simulator is based on a paper about the ["Limits of the simulation of quantum computers on classical computers"](https://journals.aps.org/prx/abstract/10.1103/PhysRevX.10.041038).
 
 ## Installation
-To install the dependencys of this project run `pip3 install numpy`. This project relies on the algebraic operations of numpy.
+To use this simulation package you need to install some dependencies.
+
+Run the following code to install them:
+
+```
+pip3 install numpy # needed for the tensor operations
+pip3 install tqdm  # displays a progress bar during running the circuit
+
+# optional for recalculating the results (plotting)
+# pip3 install matplotlib 
+```
 
 ## Examples
 

simulation/circuit.py

@@ -1,4 +1,5 @@
 import sys
+from tqdm import tqdm
 from .operations import *
 
 class Circuit:
@@ -39,7 +40,7 @@ def run(self):
         Applies the circuit.
         The Gates are applied on their qubits in the order they are in the array
         """
-        for gate in self.gates:
+        for gate in tqdm(self.gates):
             if gate["type"] == "one":
                 if self.verbose:
                     print(f'One qubit gate at {gate["i"]}')

simulation/operations.py

@@ -76,7 +76,11 @@ def calculate_fidelity(fidelities, S, chi):
 
     fidelity = np.sqrt(approx_sum) / np.sqrt(perfect_sum)
     if math.isnan(fidelity):
-        # happens if the difference is to small, therefore we approximate with 1
+        # sometime the fidelity gets "nan", this is probably the case when small values due to the square and sqrt functions are rounded up to 0
+        # hence we get a "divided by 0 -> nan"
+        # I tried to leave those values out of the calculation of the average fidelity, but that turns out to return in a weird shaped curve
+        # I got better results (well-shaped curve) when approximating this one 2-qubit gate fidelity with "1"
+        # since this happens only with small values the error, which is added here, should not be large and occurs only in this specific gate.
         fidelities.append(1)
     else:
         fidelities.append(fidelity)