Hi @<1523701087100473344:profile|SuccessfulKoala55> , thanks for the answer, I'll try that. Would you suggest any other simpler way to achieve the same result? I just want to get the best model according to a logged metric.

So in your case:

title = hashlib.md5(str("f1_score").encode("utf-8")).hexdigest()
series = hashlib.md5(str("train").encode("utf-8")).hexdigest()
Task.get_tasks(
    project_name='project',
    task_name='task',
    task_filter={'order_by': [f'-last_metrics.{title}.{series}']},
)

Hi @<1570220858075516928:profile|SlipperySheep79> , the reason it's not sorting for you is that the last metrics are not stored by the title/series name itself (as it might contain characters unsupported by the database). Instead, you need to use the format last_metrics.<md5-encoded-title>.<md5-endoed-series>.<item>
Where:

• md5 encoding is (in python): hashlib.md5(str("thestring").encode("utf-8")).hexdigest()
• and <item> can be either value , min_value or max_value

Makes sense 🙂