Reputation
Badges 1
69 × Eureka!CostlyOstrich36 I get the last iteration by task.get_last_iteration()
I want to report each iteration..
Or do I have to dive into the code in train function and write the code there?
I just added -p port_num:port_num to extra_docker_arguments
And again only in docker mode CostlyOstrich36
But now I have diffrent problem, in the dataset.get_mutable_local_copy
I get:
File "/root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages/clearml/datasets/dataset.py", line 1276, in _download_part
raise ValueError("Could not download dataset id={} entry={}".format(self._id, data_artifact_name))
ValueError: Could not download dataset
I found the problem I think, thanks!
I have nowhere else to bring iteration number ..
The sdk.aws.s3.credentials.0.host and sdk.aws.s3.credentials.0.key, yes
CostlyOstrich36 Not sure I understood, the current iterations come from the function
task.get_last_iteration() ...
Hi SuccessfulKoala55 ,
First I create the task and change the output_uri to some folder I createdtask = Task.init(project_name='main', task_name='task', output_uri=r'C:\Users\Chen\Desktop\folder)Than I used:task.upload_artifact
logger = task.get_logger()
train(model_dir=trained_model_dst, pipeline_config_path=pipeline_config_path, save_checkpoints_steps=args.checkpoints)
logger.report_scalar(title='evaluate', series='score', value=5, iteration=task.get_last_iteration())
CostlyOstrich36 tensorflow reporting it, ClearML capture it, and I get it with that function.
MelancholyElk85 So I need to change the paths each time?
Haa,
How can I change it?
The pipline is on services
and the first task on defult
I need to change to pipe.set_default_execution_queue('services')?
or leave it defult?