Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Unanswered
Hi All. What Is The Most Straightforward Way To Download All Experiment-Related Data (Configs, Scalars, Debug Samples Etc.) For One Experiment From Clearml Server (Self-Hosted)? I Want To Backup Some Chosen Experiments And Also Poke Around The Experimenta


We use task.export_task() and a hacked version to get console log:


def save_console_log(task: clearml.Task, fs, remote_path, number_of_reports=10000):
    from clearml.backend_api.services import events
    from clearml.backend_api import Session

    # Stollen from Task.get_reported_console_output()
    if Session.check_min_api_version('2.9'):
        request = events.GetTaskLogRequest(
            task=task.id,
            order='asc',
            navigate_earlier=True,
            batch_size=number_of_reports)
    else:
        request = events.GetTaskLogRequest(
            task=task.id,
            order='asc',
            from_='tail',
            batch_size=number_of_reports)

    res = task.send(request)
    response = res.wait()
    if not response.ok() or not response.response_data.get('events'):
        print("Warning: No Clearml Console log retrieved")
        return

    logs = response.response_data['events']
    json_data = json.dumps(logs, indent=2)
    print(f"Saving ClearML task console logs to {remote_path}")
    with fs.open(remote_path, "w") as json_file:
        json_file.write(json_data)
  
  
Posted 6 months ago
55 Views
0 Answers
6 months ago
6 months ago