search

IBM / AMLSim

The AMLSim project is intended to provide a multi-agent based simulator that generates synthetic banking transaction data together with a set of known money laundering patterns - mainly for the purpose of testing machine learning models and graph algorithms. We welcome you to enhance this effort since the data set related to money laundering is critical to advance detection capabilities of money laundering activities.
Apache License 2.0
251 stars 81 forks source link

tx_log.csv being created in wrong directory? #47

Open apkil opened 2 years ago

apkil commented 2 years ago

Playing with AMLSim for the first time for a dissertation project. Getting an error on this step - python3 scripts/convert_logs.py conf.json

Simulation name: sample Load alert groups: alert_members.csv Convert transaction list from tmp/sample/tx_log.csv to transactions.csv, cash_tx.csv and alert_transactions.csv Traceback (most recent call last): File "scripts/convert_logs.py", line 912, in converter.convert_acct_tx() File "scripts/convert_logs.py", line 632, in convert_acct_tx in_tx_f = open(self.log_file, "r") # Transaction log file from the Java simulator FileNotFoundError: [Errno 2] No such file or directory: 'tmp/sample/tx_log.csv'

nelsonjd commented 2 years ago

@apkil I have a pr for this currently open. You can apply the one-line fix in your local until I get it merged.

https://github.com/IBM/AMLSim/pull/46/commits/a941a07116f61f1236188e3755074c0df0e537d9

nelsonjd commented 2 years ago

@apkil This is has been fixed and merged. This issue is resolved. Thanks.

apkil commented 2 years ago

Thanks for the fix @nelsonjd. I applied the fix locally as suggested that day to move ahead. I was able to generate the data for sample simulation. Now I need to understand the param files to generate data for different typologies. Any resources to help me in that direction would be great?

apkil commented 2 years ago

Sorry to hijack this thread, but thought I would reach out for quick help. I am new to AMLSim world, I am trying to get data for my dissertation that involves Use of Machine Learning in AML. To evaluate different models what would you suggest I should use from your example dataset, something that is close to the real world and a mix of a few typologies? Would version v2.1 be suitable for that? As I am running short on time for the dissertation I have limited time to explore AMLSim in-depth to generate my own data. Would really appreciate your guidance. Thanks Kapil