Skip to main content

FAQ

General Information

Models

Experiments

Graphs and Logs

GIT and Storage

Remote Debugging (ClearML PyCharm Plugin)

Jupyter

scikit-learn

ClearML Configuration

ClearML Hosted Service

ClearML Server Deployment

ClearML Server Configuration

ClearML Server Troubleshooting

ClearML Agent

ClearML API

General Information#

How do I know a new version came out?

Starting with ClearML v0.9.3, ClearML issues a new version release notification, which appears in the log and is output to the console, when a Python experiment script is run.

For example, when a new ClearML Python Package version is available, the notification is:

CLEARML new package available: UPGRADE to vX.Y.Z is recommended!

When a new ClearML Server version is available, the notification is:

CLEARML-SERVER new version available: upgrade to vX.Y is recommended!

How do I find out ClearML version information?

ClearML server version information is available in the ClearML webapp profile page. On the bottom right of the page, it says Version, followed by three numbers: the web application version, the API server version, and the API version.

Server version information

ClearML python package information can be obtained by using pip freeze.

For example:

pip freeze|grep clearml

should return something like this:

clearml==1.0.3rc1
clearml-agent==1.0.0
clearml-session==0.3.2

Models#

How can I sort models by a certain metric?

ClearML associates models with the experiments that created them. To sort experiments by a metric, in the ClearML Web UI, add a custom column in the experiments table and sort by that metric column.


Can I store more information on the models?

Yes! For example, you can use the Task.set_model_label_enumeration method to store label enumeration:

Task.current_task().set_model_label_enumeration( {"label": int(0), } )

For more information about Task class methods, see the Task Class reference page.


Can I store the model configuration file as well?

Yes! Use the Task.set_model_config method:

Task.current_task().set_model_config("a very long text with the configuration file's content")

I am training multiple models at the same time, but I only see one of them. What happened?

Currently, in the experiment info panel, ClearML shows only the last associated model. In the ClearML Web UI, on the Projects page, the MODELS tab shows all models.

This will be improved in a future version.


Can I log input and output models manually?

Yes! Use the InputModel.import_model and Task.connect methods to manually connect an input model. Use the OutputModel.update_weights method to manually connect a model weights file.

input_model = InputModel.import_model(link_to_initial_model_file)
Task.current_task().connect(input_model)
OutputModel(Task.current_task()).update_weights(link_to_new_model_file_here)

For more information about models, see InputModel and OutputModel classes.

Experiments#

I noticed I keep getting the message "warning: uncommitted code". What does it mean?

This message is only a warning. ClearML not only detects your current repository and git commit, but also warns you if you are using uncommitted code. ClearML does this because uncommitted code means this experiment will be difficult to reproduce. You can see uncommitted changes in the ClearML Web UI, in the EXECUTION tab of the experiment info panel.

I do not use argparse for hyperparameters. Do you have a solution?

Yes! ClearML supports connecting hyperparameter dictionaries to experiments, using the Task.connect method.

For example, to log the hyperparameters learning_rate, batch_size, display_step, model_path, n_hidden_1, and n_hidden_2:

# Create a dictionary of parameters
parameters_dict = { 'learning_rate': 0.001, 'batch_size': 100, 'display_step': 1,
'model_path': "/tmp/model.ckpt", 'n_hidden_1': 256, 'n_hidden_2': 256 }
# Connect the dictionary to your CLEARML Task
parameters_dict = Task.current_task().connect(parameters_dict)

I noticed that all of my experiments appear as "Training" Are there other options?

Yes! When creating experiments and calling Task.init, you can provide an experiment type. ClearML supports multiple experiment types. For example:

task = Task.init(project_name, task_name, Task.TaskTypes.testing)

Sometimes I see experiments as running when in fact they are not. What's going on?

ClearML monitors your Python process. When the process exits properly, ClearML closes the experiment. When the process crashes and terminates abnormally, it sometimes misses the stop signal. In this case, you can safely right click the experiment in the Web-App and abort it.


My code throws an exception, but my experiment status is not "Failed". What happened?

This issue was resolved in Trains v0.9.2. Upgrade to ClearML by executing the following command:

pip install -U clearml

When I run my experiment, I get an SSL Connection error CERTIFICATE_VERIFY_FAILED. Do you have a solution?

Your firewall may be preventing the connection. Try one of the following solutions:

  • Direct python "requests" to use the enterprise certificate file by setting the OS environment variables CURL_CA_BUNDLE or REQUESTS_CA_BUNDLE. For a detailed discussion of this topic, see https://stackoverflow.com/questions/48391750/disable-python-requests-ssl-validation-for-an-imported-module.

  • Disable certificate verification

    warning

    For security reasons, it is not recommended to disable certificate verification

    1. Upgrade ClearML to the current version:

      pip install -U clearml
    2. Create a new clearml.conf configuration file (see a sample configuration file), containing:

      api { verify_certificate = False }
    3. Copy the new clearml.conf file to:

      • Linux - ~/clearml.conf
      • Mac - $HOME/clearml.conf
      • Windows - \User\<username>\clearml.conf``~/clearml.conf

How do I modify experiment names once they have been created?

An experiment's name is a user-controlled property, which can be accessed via the Task.name variable. This allows you to use meaningful naming schemes for easily filtering and comparing of experiments.

For example, to distinguish between different experiments, you can append the task ID to the task name:

task = Task.init('examples', 'train')
task.name += ' {}'.format(task.id)

Or, append the Task ID post-execution:

tasks = Task.get_tasks(project_name='examples', task_name='train')
for t in tasks:
t.name += ' {}'.format(task.id)

Another example is to append a specific hyperparameter and its value to each task's name:

tasks = Task.get_tasks(project_name='examples', task_name='my_automl_experiment')
for t in tasks:
params = t.get_parameters()
if 'my_secret_parameter' in params:
t.name += ' my_secret_parameter={}'.format(params['my_secret_parameter'])

Use this experiment naming when creating automation pipelines with a naming convention.


Using Conda and the "typing" package, I get the error "AttributeError: type object 'Callable' has no attribute '_abc_registry'". How do I fix this?

Conda and the typing package may have some compatibility issues.

However, since Python 3.5, the typing package is part of the standard library.

To resolve the error, uninstall typing and rerun you script. If this does not fix the issue, create a new ClearML issue, including the full error, and your environment details.


My ClearML Server disk space usage is too high. What can I do about this?

We designed the ClearML open source suite, including ClearML Server, to ensure experiment traceability. For this reason, the ClearML Web UI does not include a feature to delete experiments. The ClearML Web UI does allow you to archive experiments so that they appear only in the Archive area.

In rare instances, however, such as high disk usage for a privately-hosted ClearML Server because Elasticsearch is indexing unwanted experiments, you may choose to delete an experiment.

You can use the APIClient provided by ClearML Agent and client.tasks.delete() to delete an experiment.

warning

You cannot undo the deletion of an experiment.

For example, the following script deletes an experiment whose Task ID is 123456789.

from clearml_agent import APIClient
client = APIClient()
client.tasks.delete(task='123456789')

Can I change the random seed my experiment uses?

Yes! By default, ClearML initializes Tasks with a default seed. You change that seed by calling the make_deterministic method.


In the Web UI, I can't access files that my experiment stored. Why not?

ClearML stores file locations. The machine running your browser must have access to the location where the machine that ran the Task stored the file. This applies to debug samples and artifacts. If, for example, the machine running the browser does not have access, you may see "Unable to load image", instead of the image.


I get the message "CLEARML Monitor: Could not detect iteration reporting, falling back to iterations as seconds-from-start". What does it mean?

If metric reporting begins within the first three minutes, ClearML reports resource monitoring by iteration. Otherwise, it reports resource monitoring by seconds from start, and logs a message:

CLEARML Monitor: Could not detect iteration reporting, falling back to iterations as seconds-from-start.

However, if metric reporting begins after three minutes and anytime up to thirty minutes, resource monitoring reverts to by iteration, and ClearML logs a message

CLEARML Monitor: Reporting detected, reverting back to iteration based reporting.

After thirty minutes, it remains unchanged.


Can I control what ClearML automatically logs?

Yes! ClearML allows you to control automatic logging for stdout, stderr, and frameworks.

When initializing a Task by calling the Task.init method, provide the auto_connect_frameworks parameter to control framework logging, and the auto_connect_streams parameter to control stdout, stderr, and standard logging. The values are True, False, and a dictionary for fine-grain control. See Task.init.


Can I run ClearML Task while working offline?

Yes! You can use ClearML's Offline Mode, in which all the data and logs that a task captures from the code are stored in a local folder

Before initializing a task, use the Task.set_offline class method and set the offline_mode argument to True. When executed, this returns the Task ID and a path to the local session folder. In order to upload to the ClearML Server the local execution data that the Task captured offline, use the Task.import_offline_session method. See Storing Task Data Offline.

Graphs and Logs#

The first log lines are missing from the experiment console tab. Where did they go?

Due to speed/optimization issues, we opted to display only the last several hundred log lines.

You can always download the full log as a file using the ClearML Web UI. In the ClearML Web UI > experiment info panel > RESULTS tab > CONSOLE sub-tab, use the Download full log feature.


Can I create a graph comparing hyperparameters vs. model accuracy?

Yes! You can manually create a plot with a single point X-axis for the hyperparameter value, and Y-axis for the accuracy. For example:

number_layers = 10
accuracy = 0.95
Task.current_task().get_logger().report_scatter2d(
"performance", "accuracy", iteration=0,
mode='markers', scatter=[(number_layers, accuracy)])

Assuming the hyperparameter is number_layers with current value 10, and the accuracy for the trained model is 0.95. Then, the experiment comparison graph shows:

image

Another option is a histogram chart:

number_layers = 10
accuracy = 0.95
Task.current_task().get_logger().report_vector(
"performance", "accuracy", iteration=0, labels=['accuracy'],
values=[accuracy], xlabels=['number_layers %d' % number_layers])

image


I want to add more graphs, not just with TensorBoard. Is this supported?

Yes! The Logger module includes methods for explicit reporting. For examples of explicit reporting, see the Explicit Reporting tutorial, which includes a list of methods for explicit reporting.


How can I report more than one scatter 2D series on the same plot?

The Logger.report_scatter2d() method reports all series with the same title and iteration parameter values on the same plot.

For example, the following two scatter2D series are reported on the same plot, because both have a title of example_scatter and an iteration of 1:

scatter2d_1 = np.hstack((np.atleast_2d(np.arange(0, 10)).T, np.random.randint(10, size=(10, 1))))
logger.report_scatter2d("example_scatter", "series_1", iteration=1, scatter=scatter2d_1,
xaxis="title x", yaxis="title y")
scatter2d_2 = np.hstack((np.atleast_2d(np.arange(0, 10)).T, np.random.randint(10, size=(10, 1))))
logger.report_scatter2d("example_scatter", "series_2", iteration=1, scatter=scatter2d_2,
xaxis="title x", yaxis="title y")

GIT and Storage#

Is there something ClearML can do about uncommitted code running?

Yes! ClearML stores the git diff as part of the experiment's information. You can view the git diff in the ClearML Web UI, experiment info panel > EXECUTION tab.


I read there is a feature for centralized model storage. How do I use it?

When calling Task.init, providing the output_uri parameter allows you to specify the location in which model checkpoints (snapshots) will be stored.

For example, to store model checkpoints (snapshots) in /mnt/shared/folder:

task = Task.init(project_name, task_name, output_uri="/mnt/shared/folder")

ClearML will copy all stored snapshots into a subfolder under /mnt/shared/folder. The subfolder's name will contain the experiment's ID. If the experiment's ID is 6ea4f0b56d994320a713aeaf13a86d9d, the following folder will be used:

/mnt/shared/folder/task.6ea4f0b56d994320a713aeaf13a86d9d/models/

ClearML supports other storage types for output_uri, including:

# AWS S3 bucket
task = Task.init(project_name, task_name, output_uri="s3://bucket-name/folder")
# Google Cloud Storage bucket
task = Task.init(project_name, task_name, output_uri="gs://bucket-name/folder")

To use Cloud storage with ClearML, configure the storage credentials in your ~/clearml.conf. For detailed information, see ClearML Configuration Reference.


When using PyCharm to remotely debug a machine, the Git repo is not detected. Do you have a solution?

Yes! Since this is such a common occurrence, we created a PyCharm plugin that allows a remote debugger to grab your local repository / commit ID. For detailed information about using the plugin, see the ClearML PyCharm Plugin.

Jupyter#

I am using Jupyter Notebook. Is this supported?

Yes! You can run ClearML in Jupyter Notebooks using either of the following:

  • Option 1: Install ClearML on your Jupyter Notebook host machine
  • Option 2: Install ClearML in your Jupyter Notebook and connect using ClearML credentials

Option 1: Install ClearML on your Jupyter host machine

  1. Connect to your Jupyter host machine.

  2. Install the ClearML Python Package.

    pip install clearml
  3. Run the ClearML initialize wizard.

    clearml-init
  4. In your Jupyter Notebook, you can now use ClearML.

Option 2: Install ClearML in your Jupyter Notebook

  1. In the ClearML Web UI, Profile page, create credentials and copy your access key and secret key. These are required in the Step 3.

  2. Install the ClearML Python Package.

    pip install clearml
  3. Use the Task.set_credentials method to specify the host, port, access key and secret key (see step 1).

    # Set your credentials using the trains apiserver URI and port, access_key, and secret_key.
    Task.set_credentials(host='http://localhost:8008',key='<access_key>', secret='<secret_key>')
    note

    host is the API server (default port 8008), not the web server (default port 8080).

  4. You can now use ClearML.

    # create a task and start training
    task = Task.init('juptyer project', 'my notebook')

Remote Debugging (ClearML PyCharm Plugin)#

I am using your ClearML PyCharm Plugin for remote debugging. I get the message "clearml.Task - INFO - Repository and package analysis timed out (10.0 sec), giving up". What should I do?

ClearML uses a background thread to analyze the script. This includes package requirements. At the end of the execution of the script, if the background thread is still running, ClearML allows the thread another 10 seconds to complete. If the thread does not complete, it times out.

This can occur for scripts that do not import any packages, for example short test scripts.

To fix this issue, you could import the time package and add a time.sleep(20) statement to the end of your script.

scikit-learn#

Can I use ClearML with scikit-learn?

Yes! scikit-learn is supported. Everything you do is logged. ClearML automatically logs models which are stored using joblib. See the scikit-learn examples with Matplotlib and Joblib.

ClearML Configuration#

How do I explicitly specify the ClearML configuration file to be used?

To override the default configuration file location, set the CLEARML_CONFIG_FILE OS environment variable.

For example:

export CLEARML_CONFIG_FILE="/home/user/myclearml.conf"

How can I override ClearML credentials from the OS environment?

To override your configuration file / defaults, set the following OS environment variables:

export CLEARML_API_ACCESS_KEY="key_here"
export CLEARML_API_SECRET_KEY="secret_here"
export CLEARML_API_HOST="http://localhost:8008"

How can I track OS environment variables with experiments?

Set the OS environment variable ClearML_LOG_ENVIRONMENT with the variables you need track, either:

  • All environment variables:

    export ClearML_LOG_ENVIRONMENT="*"
  • Specific environment variables, for example, log PWD and PYTHONPATH:

    export ClearML_LOG_ENVIRONMENT="PWD,PYTHONPATH"
  • No environment variables:

    export ClearML_LOG_ENVIRONMENT=

ClearML Hosted Service#

I run my script, but my experiment is not in the ClearML Hosted Service Web UI. How do I fix this?

If you joined the ClearML Hosted Service and run a script, but your experiment does not appear in Web UI, you may not have configured ClearML for the hosted service. Run the ClearML setup wizard. It will request your hosted service ClearML credentials and create the ClearML configuration you need.

pip install clearml
clearml-init

ClearML Server Deployment#

How do I deploy ClearML Server on stand-alone Linux Ubuntu or macOS systems?

For detailed instructions, see Deploying ClearML Server: Linux or macOS in the "Deploying ClearML" section.


How do I deploy ClearML Server on Windows 10?

For detailed instructions, see Deploying ClearML Server: Windows 10 in the "Deploying ClearML" section.


How do I deploy ClearML Server on AWS EC2 AMIs?

For detailed instructions, see Deploying ClearML Server: AWS EC2 AMIs in the "Deploying ClearML" section.


How do I deploy ClearML Server on the Google Cloud Platform?

For detailed instructions, see Deploying ClearML Server: Google Cloud Platform in the "Deploying ClearML" section.


How do I restart ClearML Server?

For detailed instructions, see the "Restarting" section of the documentation page for your deployment format. For example, if you deployed to Linux, see Restarting on the "Deploying ClearML Server: Linux or macOS" page.


Can I deploy ClearML Server on Kubernetes clusters?

Yes! ClearML Server supports Kubernetes. For detailed instructions, see Deploying ClearML Server: Kubernetes in the "Deploying ClearML" section.


Can I create a Helm Chart for ClearML Server Kubernetes deployment?

Yes! You can create a Helm Chart of ClearML Server Kubernetes deployment. For detailed instructions, see Deploying ClearML Server: Kubernetes using Helm in the "Deploying ClearML" section.


My Docker cannot load a local host directory on SELinux?

If you are using SELinux, run the following command (see this discussion):

chcon -Rt svirt_sandbox_file_t /opt/clearml

ClearML Server Configuration#

How do I configure ClearML Server for sub-domains and load balancers?

For detailed instructions, see Configuring Sub-domains and load balancers on the "Configuring Your Own ClearML Server" page.


Can I add web login authentication to ClearML Server?

By default, anyone can login to the ClearML Server Web-App. You can configure the ClearML Server to allow only a specific set of users to access the system.

For detailed instructions, see Web Login Authentication on the "Configuring Your Own ClearML Server" page in the "Deploying ClearML" section.


Can I modify a non-responsive task settings?

The non-responsive experiment watchdog monitors experiments that were not updated for a specified time interval, and marks them as aborted. The watchdog is always active.

You can modify the following settings for the watchdog:

  • The time threshold (in seconds) of task inactivity (default value is 7200 seconds which is 2 hours).
  • The time interval (in seconds) between watchdog cycles.

For detailed instructions, see Modifying non-responsive Task watchdog settings on the "Configuring Your Own ClearML Server" page.

ClearML Server Troubleshooting#

I did a reinstall. Why can't I create credentials in the Web-App (UI)?

The issue is likely your browser cookies for ClearML Server. We recommend clearing your browser cookies for ClearML Server. For example:

  • For Firefox - go to Developer Tools > Storage > Cookies > delete all cookies under the ClearML Server URL.
  • For Chrome - Developer Tools > Application > Cookies > delete all cookies under the ClearML Server URL.

How do I fix Docker upgrade errors?

To resolve the Docker error:

... The container name "/trains-???" is already in use by ...

try removing deprecated images:

$ docker rm -f $(docker ps -a -q)

Why is web login authentication not working?

A port conflict between the ClearML Server MongoDB and / or Elastic instances, and other instances running on your system may prevent web login authentication from working correctly.

ClearML Server uses the following default ports which may be in conflict with other instances:

  • MongoDB port 27017
  • Elastic port 9200

You can check for port conflicts in the logs in /opt/clearml/log.

If a port conflict occurs, change the MongoDB and / or Elastic ports in the docker-compose.yml, and then run the Docker compose commands to restart the ClearML Server instance.

To change the MongoDB and / or Elastic ports for your ClearML Server, do the following:

  1. Edit the docker-compose.yml file.

  2. In the services/trainsserver/environment section, add the following environment variable(s):

    • For MongoDB:

      MONGODB_SERVICE_PORT: <new-mongodb-port>
    • For Elastic:

      ELASTIC_SERVICE_PORT: <new-elasticsearch-port>

      For example:

      MONGODB_SERVICE_PORT: 27018
      ELASTIC_SERVICE_PORT: 9201
  3. For MongoDB, in the services/mongo/ports section, expose the new MongoDB port:

    <new-mongodb-port>:27017

    For example:

    20718:27017
  4. For Elastic, in the services/elasticsearch/ports section, expose the new Elastic port:

    <new-elasticsearch-port>:9200

    For example:

    9201:9200
  5. Restart ClearML Server, see Restarting ClearML Server.


How do I bypass a proxy configuration to access my local ClearML Server?

A proxy server may block access to ClearML Server configured for localhost.

To fix this, you may allow bypassing of your proxy server to localhost using a system environment variable, and configure ClearML for ClearML Server using it.

Do the following:

  1. Allow bypassing of your proxy server to localhost using a system environment variable, for example:

    NO_PROXY = localhost
  2. If a ClearML configuration file (clearml.conf) exists, delete it.

  3. Open a terminal session.

  4. In the terminal session, set the system environment variable to 127.0.0.1, for example:

    • Linux:

      no_proxy=127.0.0.1
      NO_PROXY=127.0.0.1
    • Windows:

      set no_proxy=127.0.0.1
      set NO_PROXY=127.0.0.1
  5. Run the ClearML wizard clearml-init to configure ClearML for ClearML Server, which will prompt you to open the ClearML Web UI at, http://127.0.0.1:8080/, and create new ClearML credentials.

    The wizard completes with:

    Verifying credentials ...
    Credentials verified!
    New configuration stored in /home/<username>/clearml.conf
    ClearML setup completed successfully.

The ClearML Server keeps returning HTTP 500 (or 400) errors. How do I fix this?

The ClearML Server will return HTTP error responses (5XX, or 4XX) when some of its backend components are failing.

A common cause for such a failure is low available disk space, as the Elasticsearch service used by your server will go into read-only mode when it hits Elasticsearch flood watermark (by default, set to 95% disk space used).

This can be readily fixed by making more disk space available to the Elasticsearch service (either freeing up disk space disk, or if using dynamic cloud storage, increasing the disk size).

note

A likely indication of this situation can be determined by searching your clearml logs for "[FORBIDDEN/12/index read-only / allow delete (api)]".


Why is my ClearML Web-App (UI) not showing any data?

If your ClearML Web-App (UI) does not show anything, it may be an error authenticating with the server. Try clearing the application cookies for the site in your browser's developer tools.

ClearML Agent#

How can I execute ClearML Agent without installing packages each time?

Instead of installing the Python packages in the virtual environment created by ClearML Agent, you can optimize execution time by inheriting the packages from your global site-packages directory. In the ClearML configuration file, set the configuration option agent.package_manager.system_site_packages to true.

ClearML API#

How can I use the ClearML API to fetch data?

To fetch data using the ClearML API, create an authenticated session and send requests for data using the ClearML API services and methods. The responses to the requests contain your data.

For example, to get the metrics for an experiment and to print metrics as a histogram:

  1. Start an authenticated session.

  2. Send a request for all projects named examples using the projects service GetAllRequest method.

  3. From the response, get the Ids of all those projects named examples.

  4. Send a request for all experiments (tasks) with those project IDs using the tasks service GetAllRequest method.

  5. From the response, get the data for the experiment (task) ID 11 and print the experiment name.

  6. Send a request for a metrics histogram for experiment (task) ID 11 using the events service ScalarMetricsIterHistogramRequest method and print the histogram.

    # Import Session from the trains backend_api
    from trains.backend_api import Session
    # Import the services for tasks, events, and projects
    from trains.backend_api.services import tasks, events, projects
    # Create an authenticated session
    session = Session()
    # Get projects matching the project name 'examples'
    res = session.send(projects.GetAllRequest(name='examples'))
    # Get all the project Ids matching the project name 'examples"
    projects_id = [p.id for p in res.response.projects]
    print('project ids: {}'.format(projects_id))
    # Get all the experiments/tasks
    res = session.send(tasks.GetAllRequest(project=projects_id))
    # Do your work
    # For example, get the experiment whose ID is '11'
    task = res.response.tasks[11]
    print('task name: {}'.format(task.name))
    # For example, for experiment ID '11', get the experiment metric values
    res = session.send(events.ScalarMetricsIterHistogramRequest(
    task=task.id,
    ))
    scalars = res.response_data
    print('scalars {}'.format(scalars))