Skip to main content

FeersumNLU API

Project description

FeersumNLU API Wrappers

This git repo hosts Python language wrappers and cURL examples for the FeersumNLU & FeersumVision RESTful HTTP APIs. FeersumNLU & FeersumVision are the language and image comprehension components of Feersum Engine .


FeersumNLU & FeersumVision provide machine comprehension for natural chat-like interactions with your users. The components are designed to be flexible enough to work with any language, even in markets where large bodies of labelled data do not exist yet.

Our in-house comprehension models are based on the latest research and can be setup for various specialised verticals like finance and health. The APIs are also designed such that a user may add multi-lingual Natural Language Understanding (NLU) and image comprehension to their application without requiring a deep understanding of NLU and computer vision theory.

The examples in this repo demonstrate commonly required NLU and vision functions. The language wrappers are auto-generated from an API Swagger spec available at The cURL examples of direct HTTP access are in the examples_curl folder.

(Note: This repo is currently being updated with more examples. You may subscribe to receive notifications about updates at )


Current NLU features include natural language FAQ’s, detection of the user’s intent and sentiment, information extraction, entity extraction, and text-based language identification.

Current vision features include image classification and information extraction.


Please have a look at the examples in the repo. It is also recommended that you create a Python virtual environment and then follow one of the install options below to run the examples. The examples were tested with Python 3.5 and 3.6.

Creating a Python Virtual Environment

To get started with the Python wrappers install Python 3.5 (or 3.6) and pip. Then do:

$ virtualenv -p /usr/local/bin/python3.5 .pyenv
$ source .pyenv/bin/activate
$ pip install pip-tools
$ pip install appdirs

If you don’t have virtualenv installed first run:

$ pip install virtualenv
$ sudo /usr/bin/easy_install virtualenv

Install Option 1 - Using make

Clone the repo and then to install the dependencies required to run the module’s examples run:

$ make requirements
$ make deps

Install Option 2 - Using setuptools

Alternatively clone the repo and then install the feersum_nlu wrapper module into your Python environment using setuptools:

$ make requirements
$ make deps
$ python install

The benefit of installing the feersum_nlu module into you Python environment is that you can more easily use it in your own projects.

Install Option 3 - Using pip

The feersum_nlu wrapper module is also available from the Python Package Index To install it using pip run:

$ pip install feersum_nlu

The benefit of installing the feersum_nlu module into you Python environment is that you can more easily use it in your own projects.

Running the Examples

If you use an IDE like PyCharm you can simply open the folder you cloned the source to (e.g. feersum-nlu-api-wrappers) with the IDE.

Or to start executing example notebooks install Jupyter with

$ pip install jupyter

and then run:

$ jupyter notebook

The notebooks are all in the example_notebooks folder.

Alternatively run the example python scripts in the examples folder from the terminal e.g.:

$ PYTHONPATH=. python examples/

Remember to set your API token in the example scripts and notebooks.

The FeersumNLU Playground Server

A FeersumNLU playground instance of the RESTful web service is hosted at You may use this URL to run the examples given in this repo.

You’ll need an authentication token to access the service, so email us at and we’ll send you a token to use. Some autogenerated API documentation is available at

Note that this is just a playground instance. Please contact us at for a variety of hosting options including containerised solutions.

Making your Own API Wrapper

This Python language wrapper was generated using the swagger-codegen toolchain. The full process is in the makefile target called update_spec. To update the Python wrapper run:

$ make update_spec

To generate an API wrapper for another language modify the command to use a different generator. See for some more details.

To generate a PHP wrapper, for example, change the generate command to something like:

$ swagger-codegen generate -i swagger/swagger.yaml -l php -c swagger/swagger_codegen-python_config.json -o swagger/build_php


Version 2.0.46 should no longer contain the service and version postfix, for example, ‘/nlu/v2’ in ‘’. The host should now just be, for example, ‘’.

Version 2.0.45

Added the text and image labelled datasets endpoints to allow a user to create and manage labelled datasets.

Added a ‘has_labels’ flag to the model details returned on the dashboard. Useful to know which model types have a ‘get_labels’ endpoint.

Version 2.0.44

Added the approximate (fuzzy) string matching text classifier.

Added the softmax temperature as a hyper parameter to the text, intent and FAQ classifiers.

Version 2.0.43

Updated the sentiment model to add additional emoji sentiment mainly to support thumbs up/down.

Tested and fixed the emotion rest API.

Added the active log file’s name to the dashboard response to know which file to download when needed.

Version 2.0.42

Some example updates.

Changed the ‘get_labels’ rest api collection to a more resty ‘labels’. No change to wrapper functions.

Version 2.0.41

Added an SVM classifier algorithm option to the text classifier. The supported algorithms now are naive_bayes, svm, nearest_neighbour_l1, and nearest_neighbour_cosine.

Added collection_uri to the dashboard’s model detail. It is the URI of the model type’s collection resource e.g. /faq_matchers for the collection of faq_matcher models. Typically it is just the plural of the model type.

The model’s uuid is now also returned in the dashboard model detail. The model history includes the current version of the model only if the last update was pushed to the history.

The dashboard now also returns any custom sentence_encoders (language_models) and image_encoders (vision_models) that is loaded for the API key.

Version 2.0.40

Updated the ‘update_samples’ and ‘update_samples’end points to just return a list of samples.

Version 2.0.39

Added ‘update_samples’ response to update samples endpoints.

Added a samples put/update endpoint to update samples by uuid.

Version 2.0.38

Updated the sample delete endpoints to allow one to delete samples by their UUIDs. The sample UUIDs are allocated by the service and returned along with the samples from the get samples endpoints.

Version 2.0.37

Updated the API key management interface to allow one to add an entry with an existing key.

The api key details response now includes an api call breakdown over the endpoints.

Version 2.0.36

Added an emotion classifier model.

Added an image reader model for OCR and barcode reading applications.

Added optional index and len query params to the GET sample endpoints to allow one to split large sample downloads over multiple calls in cases where the service config doesn’t allow large responses.

Version 2.0.35

Added num_epochs to the image classifier’s train endpoint.

Added automatic class balancing added for the image classifier.

Updated the Python API wrapper for Python 3.7 using swagger-codegen 2.4.6.

Version 2.0.34

Server patched, no functional updates.

Version 2.0.33

Added image classifier endpoint.

Added LR4_language_recogniser alias for language_recogniser for compatibility with some older applications.

The dashboard now reports language models as sentence_encoders instead of word_manifolds. The sentence encoder model also now returns the encoder type and language.

Added examples of how to use the text classifier using Naive Bayes and nearest neighbour L1 search (with different language models).

The word manifold endpoint have been removed from the service while developing the new language model endpoints!

Add ignore word boundaries and ignore case flags to synonym extractor entities.

Version 2.0.31

Added git-like revision control. One can now load a specific revision of a model by providing its UUID in the create endpoints. The dashboard endpoint also now returns a model’s revision history.

Added a second POST dashboard end-point that allows one to set some operation parameters like a model’s history size to show.

Added a synonym entity extractor.

Added a convenience ‘delete_all’ endpoint for data objects.

Added a PrometheusHandler logging handler that logs counts of various log level emits to Prometheus.

Added X-Caller header param. E.g. x_caller=’example_caller’ in api call.

Version 2.0.29

Added TSNE analytics to the intent model and its API. See ‘examples/’.

Added model readonly flag which prevents a model from being modified or a new model of the same name being created.

Added CRF entity extractor.

Version 2.0.28

Specialised the entity extractor response models from entity_model to duckling_entity, person_name_entity, regex_entity and sim_word_entity.

Version 2.0.27

Added more samples to intent classifier to demonstrate cross validation.

Version 2.0.26

Added import and export of models to a feersum_nlu_util.transfer module.

Version 2.0.25

Added the threshold param (that used to be only on intents and FAQs) to the text classifier model.

Add num_training_samples and num_testing_samples to model instance details.

example/ added example of providing a language hint.

Version 2.0.24

Added mypy, pylint, flake8, coverage and coveralls to dependencies.

example/ updated to show how to get the model’s editable parameters (intent_classifier_get_params).

example/ updated to show how to retrieve response headers X-RateLimit-Remaining.

Version 2.0.23

Added multi-part sentiment:

  • The sentiment end-point use to return object {“value”: 0.54}.
  • Now it returns object {“detail_list”: [{“index”: 0,”len”: 20,”value”: 0.54}],”value”: 0.54}.
  • See class ‘feersum_nlu.models.sentiment.Sentiment’

Added the vaporise endpoint to permanently delete a model instance whether it is trashed or not.

Added a lang_code hint to the retrieve/inference and data endpoints.

Version 2.0.22

Name changes to make the Python API more consistent and friendlier towards code generation:

  • text_clsfr_create_details renamed to text_classifier_create_details
  • text_clsfr_instance_detail renamed to text_classifier_instance_detail
  • wm_create_details renamed to word_manifold_create_details
  • wm_instance_detail renamed to word_manifold_instance_detail
  • create_details renamed to intent_classifier_create_details for intent classifier
  • instance_detail renamed to intent_classifier_instance_detail for intent classifier
  • create_details renamed to faq_matcher_create_details for faq matcher
  • instance_detail renamed to faq_matcher_instance_detail for faq matcher
  • regex_ent_create_details renamed to regex_entity_extractor_create_details
  • regex_instance_detail renamed to regex_entity_extractor_instance_detail
  • person_name_ent_create_details renamed to person_name_entity_extractor_create_details
  • person_name_instance_detail renamed to person_name_entity_extractor_instance_detail
  • duckling_ent_create_details renamed to duckling_entity_extractor_create_details
  • duckling_instance_detail renamed to duckling_entity_extractor_instance_detail
  • sim_word_ent_create_details renamed to sim_word_entity_extractor_create_details
  • sim_word_instance_detail renamed to sim_word_entity_extractor_instance_detail
  • lr4_language_recogniser renamed to language_recogniser
  • lr4_create_details renamed to language_recogniser_create_details
  • lr4_instance_detail renamed to language_recogniser_instance_detail

Version 2.0.21

  • Examples of async training.
  • Examples of online training.

Version 2.0.20

  • Changed name of similar_entity_extractor to sim_word_entity_extractor.
  • Removed the api hit counts from the dashboard.
  • Added a trashed flag to each model in the dashboard.
  • Removed the immediate_mode arguments in the train operations.
  • The spec and Python API wrapper have been updated to rather return lists of python objects as opposed to lists of json objects.
  • Changes to LR4 to have a load from store that works like all of the other names.

Version 2.0.18

  • Added a ‘long_name’ attribute to all models. May be used as a ‘pretty’ formatted model name while the existing model name is really a slug used in urls, etc.
  • Addition of params end point to all models. Used to update and get model attributes like desc, long_name and threshold.
  • Updated LID LR4 to load from store.

Version 2.0.16

  • Added model delete endpoints.
  • Expose experimental person name entity extractor on the http API.
  • Exposed the reference_time attribute to the Duckling entity extractor.
  • Moved the examples’ and tests’ auth token and host config to a central location in the __init.py__
  • Added this changelog.

Project details

Download files

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

Files for feersum_nlu, version 2.0.46
Filename, size File type Python version Upload date Hashes
Filename, size feersum_nlu-2.0.46.tar.gz (142.8 kB) File type Source Python version None Upload date Hashes View

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