mindmeld.components package

This module contains the components of the MindMeld platform

class mindmeld.components.Conversation(app=None, app_path=None, nlp=None, context=None, default_params=None, force_sync=False, verbose=False)

Bases: object

The conversation object is a very basic MindMeld client.

It can be useful for testing out dialogue flows in python.

Example

>>> convo = Conversation(app_path='path/to/my/app')
>>> convo.say('Hello')
['Hello. I can help you find store hours. How can I help?']
>>> convo.say('Is the store on elm open?')
['The 23 Elm Street Kwik-E-Mart is open from 7:00 to 19:00.']

history

list -- The history of the conversation. Starts with the most recent message.

context

dict -- The context of the conversation, containing user context.

default_params

Params -- The default params to use with each turn. These defaults will be overridden by params passed for each turn.

params

FrozenParams -- The params returned by the most recent turn.

force_sync

bool -- Force synchronous return for say() and process() even when app is in async mode.

verbose

bool, optional -- If True, returns class probabilities along with class prediction.

process(text, params=None, force_sync=False)

Send a message in the conversation. The message will be processed by the app based on the current state of the conversation and returns the response.

Parameters:

• text (str) -- The text of a message.
• params (dict) -- The params to use with this message, overriding any defaults which may have been set.
• force_sync (bool, optional) -- Force synchronous response even when app is in async mode.

Returns:

The dictionary response.

Return type:

(dict)

reset()

Reset the history, frame and params of the Conversation object.

say(text, params=None, force_sync=False)

Send a message in the conversation. The message will be processed by the app based on the current state of the conversation and returns the extracted messages from the directives.

Parameters:

• text (str) -- The text of a message.
• params (dict) -- The params to use with this message, overriding any defaults which may have been set.
• force_sync (bool, optional) -- Force synchronous response even when app is in async mode.

Returns:

A text representation of the dialogue responses.

Return type:

(list)

class mindmeld.components.CustomAction(name: str, config: dict, merge: bool = True)

Bases: object

This class allows the client to send Request and Responder to another server and return the

the directives and frame in the response.

get_json_payload(request, responder)

invoke(request, responder, async_mode=False)

Invoke the custom action with Request and Responder and return True if the action is executed successfully, False otherwise. Upon successful execution, we update the Frame and Directives of the Responder object.

Parameters:

• request (Request) --
• responder (DialogueResponder) --
• async_mode (bool) --

Returns:

(bool)

invoke_async(request, responder)

Asynchronously invoke the custom action with Request and Responder and return True if the action is executed successfully, False otherwise. Upon successful execution, we update the Frame and Directives of the Responder object.

Parameters:

• request (Request) --
• responder (DialogueResponder) --

Returns:

(bool)

post(json_data)

post_async(json_data)

class mindmeld.components.CustomActionSequence(actions, config, merge=True)

Bases: object

This class implements a sequence of custom actions

invoke(request, responder)

invoke_async(request, responder)

class mindmeld.components.DialogueResponder(frame=None, params=None, history=None, slots=None, request=None, dialogue_state=None, directives=None, form=None)

Bases: object

The dialogue responder helps generate directives and fill slots in the system-generated natural language responses.

class DirectiveNames

Bases: object

The list of directive names.

LIST = 'list'

LISTEN = 'listen'

REPLY = 'reply'

RESET = 'reset'

SLEEP = 'sleep'

SPEAK = 'speak'

SUGGESTIONS = 'suggestions'

class DirectiveTypes

Bases: object

The list of directive types.

ACTION = 'action'

VIEW = 'view'

act(name, payload=None)

Adds an arbitrary directive of type 'action'.

Parameters:

• name (str) -- The name of the directive.
• payload (dict, optional) -- The payload for the action.

direct(name, dtype, payload=None)

Adds an arbitrary directive.

Parameters:

• name (str) -- The name of the directive.
• dtype (str) -- The type of the directive.
• payload (dict, optional) -- The payload for the view.

display(name, payload=None)

Adds an arbitrary directive of type 'view'.

Parameters:

• name (str) -- The name of the directive.
• payload (dict, optional) -- The payload for the view.

exit_flow()

Exit the current flow by clearing the target dialogue state.

list(items)

Adds a 'list' view directive.

Parameters:items (list) -- The list of dictionary objects.

listen()

Adds a 'listen' directive.

prompt(text)

Alias for reply(). Deprecated.

Parameters:text (str) -- The text of the reply.

reply(text)

Adds a 'reply' directive.

Parameters:text (str) -- The text of the reply.

reset()

Adds a 'reset' directive.

respond(directive)

Adds an arbitrary directive.

Parameters:directive (dict) -- A directive.

sleep(delay=0)

Adds a 'sleep' directive.

Parameters:delay (int) -- The amount of milliseconds to wait before putting the client to sleep.

speak(text)

Adds a 'speak' directive.

Parameters:text (str) -- The text to speak aloud.

suggest(suggestions)

Adds a 'suggestions' directive.

Parameters:suggestions (list) -- A list of suggestions.

history

params

request

class mindmeld.components.DialogueManager(responder_class=None, async_mode=False)

Bases: object

add_dialogue_rule(name, handler, **kwargs)

Adds a dialogue state rule for the dialogue manager.

Parameters:

• name (str) -- The name of the dialogue state.
• handler (function) -- The dialogue state handler function.
• kwargs (dict) -- A list of options to be passed to the DialogueStateRule initializer.

add_middleware(middleware)

Adds middleware for the dialogue manager. Middleware will be called for each message before the dialogue state handler. Middleware registered first will be called first.

Parameters:middleware (callable) -- A dialogue manager middleware function.

apply_handler(request, responder, target_dialogue_state=None)

Applies the dialogue state handler for the most complex matching rule.

Parameters:

• request (Request) -- The request object.
• responder (DialogueResponder) -- The responder object.
• target_dialogue_state (str, optional) -- The target dialogue state.

Returns:

A DialogueResponder containing the dialogue state and directives.

Return type:

(DialogueResponder)

handle(**kwargs)

A decorator that is used to register dialogue state rules.

middleware(*args)

A decorator that is used to register dialogue handler middleware.

static reprocess(target_dialogue_state=None)

Forces the dialogue manager to back out of the flow based on the initial target dialogue state setting and reselect a handler, following a new target dialogue state

Parameters:target_dialogue_state (str, optional) -- a dialogue_state name to push system into

logger = <Logger mindmeld.components.dialogue.DialogueManager (WARNING)>

class mindmeld.components.NaturalLanguageProcessor(app_path, resource_loader=None, config=None, progress_bar=None)

Bases: mindmeld.components.nlp.Processor

The natural language processor is the MindMeld component responsible for understanding the user input using a hierarchy of natural language processing models.

domain_classifier

DomainClassifier -- The domain classifier for this application.

extract_nlp_masked_components_list(allow_nlp_components_list=None, deny_nlp_components_list=None)

This function validates a user inputted list of allowed nlp components against the NLP hierarchy and construct a hierarchy dictionary as follows: {domain: {intent: {}} if the validation of list of allowed nlp components has passed.

Parameters:

• allow_nlp_components_list (list) -- A list of allow NLP components in the format "domain.intent.entity.role".
• deny_nlp_components_list (list) -- A list of deny NLP components in the format "domain.intent.entity.role".

Returns:

A dictionary of NLP hierarchy.

Return type:

(dict)

inspect(markup, domain=None, intent=None, dynamic_resource=None)

Inspect the marked up query and print the table of features and weights.

Parameters:

• markup (str) -- The marked up query string.
• domain (str) -- The gold value for domain classification.
• intent (str) -- The gold value for intent classification.
• dynamic_resource (dict, optional) -- A dynamic resource to aid NLP inference.

static print_inspect_stats(stats)

Prints formatted output matrix

process(query_text, allowed_nlp_classes=None, allowed_intents=None, allow_nlp=None, deny_nlp=None, locale=None, language=None, time_zone=None, timestamp=None, dynamic_resource=None, verbose=False)

Processes the given query using the full hierarchy of natural language processing models trained for this application.

Parameters:

• query_text (str, tuple) -- The raw user text input, or a list of the n-best query transcripts from ASR.
• allowed_nlp_classes (dict, optional) -- A dictionary of the NLP hierarchy that is selected for NLP analysis. An example: {'smart_home': {'close_door': {}}} where smart_home is the domain and close_door is the intent.
• allowed_intents (list, optional) -- A list of allowed intents to use for the NLP processing.
• allow_nlp (list, optional) -- A list of allow NLP components to use for the NLP processing.
• deny_nlp (list, optional) -- A list of denied NLP components to use for the NLP processing.
• locale (str, optional) -- The locale representing the ISO 639-1 language code and ISO3166 alpha 2 country code separated by an underscore character.
• language (str, optional) -- Language as specified using a 639-1/2 code. This parameter is ignored deprecated this is an application level parameter.
• time_zone (str, optional) -- The name of an IANA time zone, such as 'America/Los_Angeles', or 'Asia/Kolkata' See the [tz database](https://www.iana.org/time-zones) for more information.
• timestamp (long, optional) -- A unix time stamp for the request (in seconds).
• dynamic_resource (dict, optional) -- A dynamic resource to aid NLP inference.
• verbose (bool, optional) -- If True, returns class probabilities along with class prediction.

Returns:

A processed query object that contains the prediction results from applying the full hierarchy of natural language processing models to the input query.

Return type:

(ProcessedQuery)

process_query(query, allowed_nlp_classes=None, dynamic_resource=None, verbose=False)

Processes the given query using the full hierarchy of natural language processing models trained for this application.

Parameters:

• query (Query, tuple) -- The user input query, or a list of the n-best transcripts query objects.
• allowed_nlp_classes (dict, optional) -- A dictionary of the NLP hierarchy that is selected for NLP analysis. An example: {'smart_home': {'close_door': {}}} where smart_home is the domain and close_door is the intent. If allowed_nlp_classes is None, we just use the normal model predict functionality.
• dynamic_resource (dict, optional) -- A dynamic resource to aid NLP inference.
• verbose (bool, optional) -- If True, returns class probabilities along with class prediction.

Returns:

A processed query object that contains the prediction results from applying the full hierarchy of natural language processing models to the input query.

Return type:

(ProcessedQuery)

unload()

domains

The domains supported by this application.

class mindmeld.components.QuestionAnswerer

Bases: object

Backwards compatible QuestionAnswerer class

old usages (allowed but will soon be deprecated)

# loading KB directly through class method >>> QuestionAnswerer.load_kb(...) # instantiating a QA object from QuestionAnswerer instead of QuestionAnswererFactory >>> question_answerer = QuestionAnswerer(app_path, resource_loader, es_host, config)

new usages

>>> question_answerer = QuestionAnswererFactory.create_question_answerer(**kwargs)
# Use the QA object's methods to load KB and get search results, instead of class methods
>>> question_answerer.load_kb(...)
>>> question_answerer.get(...) # .get(...) and .build_search(...)

classmethod load_kb(app_namespace, index_name, data_file, es_host=None, es_client=None, connect_timeout=2, clean=False, app_path=None, config=None, **kwargs)

Implemented to maintain backward compatibility. Should be removed in future versions.

Parameters:

• app_namespace (str) -- The namespace of the app. Used to prevent collisions between the indices of this app and those of other apps.
• index_name (str) -- The name of the new index to be created.
• data_file (str) -- The path to the data file containing the documents to be imported into the knowledge base index. It could be either json or jsonl file.
• es_host (str) -- The Elasticsearch host server.
• es_client (Elasticsearch) -- The Elasticsearch client.
• connect_timeout (int, optional) -- The amount of time for a connection to the Elasticsearch host.
• clean (bool) -- Set to true if you want to delete an existing index and reindex it
• app_path (str) -- The path to the directory containing the app's data
• config (dict) -- The QA config if passed directly rather than loaded from the app config

DEPRECATION_MESSAGE = "Calling QuestionAnswerer class directly will be deprecated in future versions. To instantiate a QA instance, use the QuestionAnswererFactory by calling 'qa = QuestionAnswererFactory.create_question_answerer(**kwargs)'. An instantiated QA can then be used as 'qa.load_kb(...)', 'qa.get(...)', etc. See https://www.mindmeld.com/docs/userguide/kb.html for details about the various functionalities available with different question-answerers."

class mindmeld.components.QuestionAnswererFactory

Bases: object

Factory class for creating QuestionAnswerers

usage

>>> question_answerer = QuestionAnswererFactory.create_question_answerer(**kwargs)
>>> question_answerer.load_kb(...)
>>> question_answerer.get(...) # .get(...) or .build_search(...)

classmethod create_question_answerer(app_path=None, config=None, app_namespace=None, **kwargs)

Parameters:

• app_path (str, optional) -- The path to the directory containing the app's data. If provided, used to obtain default 'app_namespace' and QA configurations
• app_namespace (str, optional) -- The namespace of the app. Used to prevent collisions between the indices of this app and those of other apps.
• config (dict, optional) -- The QA config if passed directly rather than loaded from the app config

class mindmeld.components.EntityResolver

Bases: object

Class for backwards compatibility

deprecated usage

>>> entity_resolver = EntityResolver(
        app_path, resource_loader, entity_type
    )

new usage

>>> entity_resolver = EntityResolverFactory.create_resolver(
        app_path, entity_type
    )
# or ...
>>> entity_resolver = EntityResolverFactory.create_resolver(
        app_path, entity_type, resource_loader=resource_loader
    )

class mindmeld.components.EntityResolverFactory

Bases: object

classmethod create_resolver(app_path, entity_type, config=None, resource_loader=None, **kwargs)

Identifies appropriate entity resolver based on input config and

returns it.

Parameters:

• app_path (str) -- The application path.
• entity_type (str) -- The entity type associated with this entity resolver.
• resource_loader (ResourceLoader) -- An object which can load resources for the resolver.
• er_config (dict) -- A classifier config
• es_host (str) -- The Elasticsearch host server.
• es_client (Elasticsearch) -- The Elasticsearch client.

class mindmeld.components.Request(*, domain=None, intent=None, entities=NOTHING, history=NOTHING, text=None, frame=immutables.Map({}), params=FrozenParams(allowed_intents=(), allow_nlp=(), deny_nlp=(), target_dialogue_state=None, time_zone=None, timestamp=None, language=None, locale=None, dynamic_resource=immutables.Map({})), context=immutables.Map({}), confidences=immutables.Map({}), nbest_transcripts_text=NOTHING, nbest_transcripts_entities=NOTHING, nbest_aligned_entities=NOTHING, form=NOTHING)

Bases: object

The Request is an object passed in through the Dialogue Manager and contains all the information provided by the application client for the dialogue handler to act on. Note: the Request object is read-only since it represents the client state, which should not be mutated.

domains

str -- Domain of the current query.

intent

str -- Intent of the current query.

entities

list of dicts -- A list of entities in the current query.

history

list of dicts -- List of previous and current responder objects (de-serialized) up to the current conversation.

text

str -- The query text.

frame

Immutables Map of stored data across multiple dialogue turns.

params

Params -- An object that modifies how MindMeld process the current turn.

context

dict -- Immutables Map containing front-end client state that is passed to the application from the client in the request.

confidences

dict -- Immutables Map of keys domains, intents, entities and roles containing confidence probabilities across all labels for each classifier.

nbest_transcripts_text

tuple -- List of alternate n-best transcripts from an ASR system

nbest_transcripts_entities

tuple -- List of lists of extracted entities for each of the n-best transcripts.

nbest_aligned_entities

tuple -- List of lists of aligned entities for each of the n-best transcripts.

mindmeld.components.invoke_custom_action(name, config, request, responder, merge=True)

mindmeld.components.invoke_custom_action_async(name, config, request, responder, merge=True)

Submodules

• mindmeld.components.classifier module
• mindmeld.components.client module
• mindmeld.components.custom_action module
• mindmeld.components.dialogue module
• mindmeld.components.domain_classifier module
• mindmeld.components.entity_recognizer module
• mindmeld.components.entity_resolver module
• mindmeld.components.intent_classifier module
• mindmeld.components.nlp module
• mindmeld.components.parser module
• mindmeld.components.question_answerer module
• mindmeld.components.request module
• mindmeld.components.role_classifier module
• mindmeld.components.schemas module
• mindmeld.components.translators module