Some troubles in Qiskit version 0.19.1

[pearcandy]

I met the following erros when executing tutorial.ipynb in Qiskit 0.19.1.

example2 = pattern.recognition( circuit=circ, level=2, n_patterns=4, min_num_nodes=4, max_num_nodes=8, min_n_repetition=2)

TypeError Traceback (most recent call last)
/tmp/ipykernel_51513/119591637.py in
----> 1 example2 = pattern.recognition( circuit=circ, level=2, n_patterns=4, min_num_nodes=4, max_num_nodes=8, min_n_repetition=2)
:
:
147 for edge in dag.edges(node):
148 child = edge[1]
--> 149 qbit = edge[2]['wire']
150 if child in node_list:
151 G.add_edge(node, child, edge_attr=(str(qbit.index)))

TypeError: 'Qubit' object is not subscriptable

from aqcel.optimization import slim

ImportError Traceback (most recent call last)
/tmp/ipykernel_51513/2067089768.py in
----> 1 from aqcel.optimization import slim
:
:
1 import numpy as np
----> 2 from aqcel.optimization import RIIM_tools
3 from qiskit import *
4 from qiskit.converters import circuit_to_dag, dag_to_circuit
5 from qiskit.dagcircuit import DAGCircuit

ImportError: cannot import name 'RIIM_tools' from 'aqcel.optimization' (../aqcel/optimization/init.py)

I found some bugs and tried to modify the following points.
However, I could not reproduce the results in ./demo/tutorial.ipynb, circuit optimization level1, 2.

・remove 'wire' key for Qubit object
149th line and 156th line in /aqcel/optimization/pettern.py

#qbit = edge[2]['wire']
qbit = edge[2]

・comment out unused library
2nd line in /aqcel/optimization/slim.py

#from aqcel.optimization import RIIM_tools

・fix an undefined variable 'x' in gate loop
223th line in /aqcel/optimization/slim.py

#if (gate[1][x].register == delete_register) & (gate[1][x].index == delete_index):
if (gate[1][0].register == delete_register) & (gate[1][0].index == delete_index):

My qiskit version is
{'qiskit-terra': '0.19.1', 'qiskit-aer': '0.10.2', 'qiskit-ignis': '0.7.0', 'qiskit-ibmq-provider': '0.18.3',
'qiskit-aqua': '0.9.5', 'qiskit': '0.34.1', 'qiskit-nature': '0.3.0', 'qiskit-finance': '0.3.0',
'qiskit-optimization': '0.3.0', 'qiskit-machine-learning': '0.3.0'}.

[WonhoJang]

I am very sorry for the late reply, @pearcandy. Actually we will stop supporting quantum gate pattern recognition for a while. About the errors on circuit optimizations, try a new code and if you find similar bugs, please let me know.

Sincerely.