Skip to main content

Python client for Fiddler Service

Project description

Fiddler Client

Python client for interacting with Fiddler. Provides a user-friendly interface to our REST API and enables event publishing for use with our monitoring features.


Requires Python >= Python-3.6.3 and pip >= 19.0


$ pip3 install fiddler-client

API Example Usage

Documentation for the API can be found here. For examples of interacting with our APIs, please check out our Quick Start Guide as well as the work notebooks found on our Samples Github.

Version History



  • New Features

    • New methods addition for alert rules: add_alert_rule, get_alert_rules, delete_alert_rule
    • New method to get triggered alerts for an alert rule: get_triggered_alerts


  • Modifications

    • Assert nullable columns in missing_value_encodings(If users send non-nullable columns as missing_value_encodings, Fiddler converts them as nullable automatically with a warning)


  • Modifications

    • Allow types other than Column data_type for missing_value_encodings.


  • Modifications

    • Accept string 'inf' in float columns in missing_value_encodings.


  • New Features

    • Support missing_value_encodings as a new field of model_info object.


  • Modifications

    • Minor bug fix to handle string nan


  • Modifications

    • Default client initiation is now the v2 client
    • publish_events_batch is now async, returns status id and doesn't wait for the upload to complete.
    • Default behavior of all publish data in v2 client is async (is_sync = False)


  • New Features

    • New capabilities for Artifact-less Monitoring


  • Modifications

    • Change the batch_size argument default to 1000 for trigger_pre_computation
    • Updated the delete_model API default value for the delete_prod parameter from False to True.
      • We will by default delete all the events associated with the model.


  • Modifications

    • Added check for "model" key before access in from_dict
    • Allow changing artifact_status when updating the model
    • Adds docstrings for add_model, add_model_surrogate and add_model_artifact


  • Modifications

    • Fixed publish_events_batch_schema backward compatible.


  • Modifications

    • Added add_model_surrogate and add_model_artifact APIs for artifactless monitoring
    • Simplifies the add_model API by removing unnecessary parameters
    • Fixed publish_events_batch_schema parameter names.


  • Modifications

    • Fixed a type coercion bug that caused some get_slice calls to fail cryptically


  • Modifications

    • Map Tree shap values from log odds space to probability space
    • Added add_model API for artifactless monitoring
    • Fixed bug in request when creating a model using add_model


  • Modifications

    • Fixed a bug that prevented importing the client in some environments.


  • Modifications

    • Removed unnecessary server-client version check that produced an uninformative warning.


  • New Features

    • New WeightingParams object. This enables weighted histograms for class-imbalanced models.
  • Modifications

    • update_model allows some small modifications in model info for the following fields: custom_explanation_names, preferred_explanation_method, display_name, description, framework, algorithm and model_deployment_params


  • New Features

    • Add v2 client. v2 methods can be accessed either via sub-module (client.v2.) or by instantiating the FiddlerApi and passing version=2.
  • Modifications

    • Remove handlers from root logger
    • Add url, org_id, auth_token and version validation while instantiating client
    • Fix dataset ingestion file extension issue
    • init monitoring issue
    • Fix publish_event request header bug
    • Add publish_events_batch_dataframe and upload_dataset_dataframe methods
    • Support for DatasetInfo class
    • Using http_client package. A wrapper over requetsts.


  • Modifications

    • Add client v2 sub-package.


  • Modifications
    • relax the version requirements for requests.
    • adds flag to init_monitoring to enable synchronous initialization


  • Modifications
    • Fixed the JSON transformation issue which was forcing requests package upgrade issue


  • New Features
    • Tree SHAP Helper.
  • Modifications
    • fdl.ModelInfo has an additional optional parameter to enabled Tree Shap


  • New Features
    • Integrated Gradients Keras TF2 Helpers.
  • Modifications
    • Relax botocore version requirements.


  • Modifications
    • Minor bug fixes and improvements.
    • run_explanation has two additional optional arguments (n_permutation and n_background) allowing users to change the default parameters for Fiddler SHAP explanations.


Inaugural client for Fiddler 22.0! This version includes numerous improvements for stability, performance, and usability.

Compatible with server versions >=22.0.0.

  • Modifications
    • Minor bug fixes and improvements.

  • Modifications
    • Minor bug fixes and improvements.

  • Modifications
    • Add a parameter in list_projects API to get detailed project information
    • Minor bug fix for the datetime format.

  • Modifications
    • Add a parameter in list_projects API to get detailed project information
    • Allow run_explanation api call to pass a list of explanation with ig_flex and one of the shap algorithm

  • Modifications
    • Minor bugfix for categorical feature drift

  • Modifications
    • Addressed an issue with categorical features with string literals containing numeric content

  • Modifications
    • Implement a ranking surrogate model for ranking task models

  • Modifications
    • change the dependecy of requests package to 0.25.1


  • Modifications
    • Improved SegmentInfo validation.
    • make the dependency versions less strict.


  • New Features
    • New publish_events_batch_schema API call, Publishes a batch events object to Fiddler Service using the passed publish_schema as a template.
    • New Ranking Monitoring capability available with publish_events_batch API
  • Modifications
    • Enforced package versions in
    • trigger_pre_computation has an additional optional argument (cache_dataset) to enable/disable dataset histograms caching.
    • register_model has 3 additional optional arguments to enable/disable pdp caching (set to False by default), feature importance caching (set to True by default) and dataset histograms caching (set to True by default).


  • New Features
    • New segment monitoring related functionality (currently in preview):
      • Ability to create and validate SegmentInfo objects,
      • upload_segment BE call,
      • activate_segment BE call,
      • deactivate_segment BE call, and
      • list_segments BE call,
  • Modifications
    • Upon connecting to the server, the client now performs a version check for the server by default. Earlier the default was to only do a version check for the client.


  • New Features
    • New update_event parameter for publish_events_batch API.
    • Changes to fdl.publish_event():
      • Renamed parameter event_time_stamp to event_timestamp
      • Added new parameter: timestamp_format
        • Allows specification of timestamp format using the FiddlerTimestamp class


  • New Features
    • New initialize_monitoring API call, sets up monitoring for a model. Intended to also work retroactively for legacy schema.
  • Modifications
    • Modified DatasetInfo.from_dataframe and ModelInfo.from_dataset_info to take additional dataset_id as parameter.
    • Modified the outputs parameter of ModelInfo.from_dataset_info to now expect a dictionary in case of regression tasks, specifying output range.
    • Modified the preferred_explanation_method parameter of ModelInfo.from_dataset_info to accept string names from custom_explanation_names. Details in docstring.
    • Misc bug fixes and documentation enhancements.


  • New Features
    • Changed the default display for ModelInfo and DatasetInfo to render HTML instead of plaintext, when accessed via jupyter notebooks
    • Added support for GCP Storage ingestion of log events using fdl.BatchPublishType.GCP_STORAGE


  • New Features
    • Restructured the following arguments for fdl.ModelInfo.from_dataset_info()
      • Added: categorical_target_class_details:
        • Mandatory for Multiclass classification tasks, optional for Binary (unused for Regression)
        • Used to specify the positive class for Binary classification, and the class order for Multiclass classification
      • Modified: target:
        • No longer optional, models must specify target columns


  • New Features
    • Restructured the following arguments for fdl.publish_events_batch()
      • Added: id_field:
        • Column to extract id value from
      • Added: timestamp_format:
        • Format of timestamp within batch object. Can be one of:
          • fdl.FiddlerTimestamp.INFER
          • fdl.FiddlerTimestamp.EPOCH_MILLISECONDS
          • fdl.FiddlerTimestamp.EPOCH_SECONDS
          • fdl.FiddlerTimestamp.ISO_8601 Removed: default_timestamp
    • Minor bug fixes
  • Deprecation Warning
    • Support fdl.publish_events_log and fdl.publish_parquet_s3 will soon be deprecated in favor of fdl.publish_events_batch()


  • Dataset Refactor
    • Datasets refactored to be members of a Project
      • This is a change promoting Datasets to be first class within Fiddler. It will affects both the UI and several API in Fiddler
    • Many API utilizing Projects will now require project_id passed as a parameter
  • New Features
    • Added fdl.update_model() to client
      • update the specified model, with model binary and from the specified model_dir
    • Added fdl.get_model() to client
      • download the model binary, and model.yaml to the given output dir.
    • Added fdl.publish_events_batch() to client
      • Publishes a batch events object to Fiddler Service.
      • Note: Support for other batch methods including fdl.publish_events_log() and fdl.publish_parquet_s3() will be deprecated in the near future in favor of fdl.publish_events_batch()
  • Changes
    • Simplified logic within fld.upload-dataset()
    • Added client/server handshake for checking version compatibilities
      • Warning issued in case of mismatch
    • Deleted redundant APIs
      • fdl.create_surrogate_model()
      • fdl.upload_model_sklearn()
    • Restructured APIs to be more duck typing-friendly (relaxing data type restrictions)
    • Patches for minor bug-fixes


  • Features
    • Minor updates to ease use of binary classification labels


  • Features
    • Added new arguments to ModelInfo.from_dataset_info()
      • preferred_explanation_method to express a preferred default explanation algorithm for a model
      • custom_explanation_names to support user-provided explanation algorithms which the user will implement on their model object via


  • Features
    • Minor improvements to publish_events_log() to circumvent datatype conversion issues


  • Features
    • Added strict name checks


  • Features
    • Added client-native multithreading support for publish_events_log() using new parameters num_threads and batch_size


  • Features
    • Added fdl.generate_sample_events() to client
      • API for generating monitoring traffic to test out Fiddler
    • Added fdl.trigger_pre_computation() to client
      • Triggers various precomputation steps within the Fiddler service based on input parameters.
    • Optionally add proxies to FiddlerApi() init


  • Features
    • Added fdl.publish_parquet_s3() to client
      • Publishes parquet events file from S3 to Fiddler instance. Experimental and may be expanded in the future.


  • Features
    • Added fdl.register_model() to client
      • Register a model in fiddler. This will generate a surrogate model, which can be replaced later with original model.

Project details

Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fiddler-client-1.5.0.tar.gz (146.0 kB view hashes)

Uploaded source

Built Distribution

fiddler_client-1.5.0-py3-none-any.whl (167.3 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page