Commit bbc8ed70 authored by Carel van Niekerk's avatar Carel van Niekerk
Browse files

Add Docs for website

parent 35da1af9
****************
Acknowledgements
****************
The following current and former group members have contributed to PyDial (in alphabetical order):
- Paweł Budzianowski
- Iñigo Casanueva
- Milica Gašić
- Dongho Kim
- Nikola Mrkšić
- Lina Rojas-Barahona
- Stefan Ultes
- Pei-Hao Su
- David Vandyke
- Gellert Weisz
- Tsung-Hsien Wen
- Steve Young
Agent - the Complete Dialogue System
************************************
.. |.| raw:: html
<br />
.. automodule:: Agent
:members:
Belief Tracking
***********************
.. |.| raw:: html
<br />
.. automodule:: belieftracking.BeliefTrackingManager
:members:
:private-members:
.. automodule:: belieftracking.BeliefTracker
:members:
:private-members:
.. automodule:: belieftracking.baseline
.. autoclass:: belieftracking.baseline.BaselineTracker
.. autoclass:: belieftracking.baseline.FocusTracker
.. automodule:: belieftracking.BeliefTrackingUtils
:members:
Conversational Entity Dialogue Model
************************************
The Conversational Entity Dialogue Model (CEDM) aims at resolving restrictions of conventional single- or multi-domain dialogue models in their capabilities of modelling complex dialogue structures, e.g., relations. The CEDM is a novel dialogue model that is centred around entities and is able to model relations as well as multiple entities of the same type `(Ultes et al., 2018)
<http://www.pydial.org/cedm>`_. A prototype of the CEDM has been implemented and integrated into the `PyDial <http://www.pydial.org/>`_ toolkit.
Please note that the CEDM prototype implementation has been designed in a way to exploit structures and implementations of PyDial where possible. Thus, some restrictions needed to be posed on the implementation which need to be resolved in future versions.
Usage Instructions
------------------
The CEDM is disabled by default. To activate and use it, download the latest version of PyDial. From Pydial root, run the activation shell script
.. code-block:: shell
python cedm/activation/activate.py
which will copy all necessary files to their respective locations. You will now find a SimulateCEDM.py in the root folder of PyDial which you can use to run experiments with the CEDM. To test it, run
.. code-block:: shell
python SimulateCEDM.py -c config/cedm_example.cfg
Evaluation
***************
.. |.| raw:: html
<br />
.. automodule:: evaluation.EvaluationManager
:members:
:private-members:
.. automodule:: evaluation.SuccessEvaluator
.. autoclass:: evaluation.SuccessEvaluator.ObjectiveSuccessEvaluator
.. autoclass:: evaluation.SuccessEvaluator.SubjectiveSuccessEvaluator
Natural Language Generation
***************************
.. |.| raw:: html
<br />
.. automodule:: semo.SemOManager
:members:
:private-members:
.. automodule:: semo.RuleSemOMethods
.. autoclass:: semo.RuleSemOMethods.PassthroughSemO
.. autoclass:: semo.RuleSemOMethods.BasicSemO
.. autoclass:: semo.RuleSemOMethods.BasicTemplateRule
.. autoclass:: semo.RuleSemOMethods.BasicTemplateFunction
.. autoclass:: semo.RuleSemOMethods.BasicTemplateFunctionRule
.. autoclass:: semo.RuleSemOMethods.BasicTemplateGenerator
.. automodule:: semo.RNNSemOMethods
.. autoclass:: semo.RNNSemOMethods.RNNSemO
Ontology
**********
.. |.| raw:: html
<br />
.. automodule:: ontology.DataBase
:members:
.. automodule:: ontology.FlatOntologyManager
:members:
.. automodule:: ontology.Ontology
:members:
.. automodule:: ontology.OntologyUtils
:members:
Policy
*********************
.. |.| raw:: html
<br />
.. automodule:: policy.Policy
:members:
:private-members:
.. automodule:: policy.PolicyManager
:members:
:private-members:
.. automodule:: policy.PolicyCommittee
:members:
:private-members:
.. automodule:: policy.HDCPolicy
.. autoclass:: policy.HDCPolicy.HDCPolicy
.. automodule:: policy.GPPolicy
.. autoclass:: policy.GPPolicy.GPPolicy
.. autoclass:: policy.GPPolicy.Kernel
.. autoclass:: policy.GPPolicy.GPAction
.. autoclass:: policy.GPPolicy.GPState
.. autoclass:: policy.GPPolicy.TerminalGPAction
.. autoclass:: policy.GPPolicy.TerminalGPState
.. automodule:: policy.GPLib
.. autoclass:: policy.GPLib.GPSARSA
.. autoclass:: policy.GPLib.GPSARSAPrior
.. autoclass:: policy.GPLib.LearnerInterface
.. automodule:: policy.HDCTopicManager
.. autoclass:: policy.HDCTopicManager.HDCTopicManagerPolicy
.. automodule:: policy.WikipediaTools
.. autoclass:: policy.WikipediaTools.WikipediaDM
.. automodule:: policy.SummaryAction
.. autoclass:: policy.SummaryAction.SummaryAction
.. automodule:: policy.SummaryUtils
.. automodule:: policy.PolicyUtils
:members:
.. automodule:: policy.BCM_Tools
DeepRL Policies
*********************
.. automodule:: policy.A2CPolicy
.. automodule:: policy.ACERPolicy
.. automodule:: policy.BDQNPolicy
.. automodule:: policy.DQNPolicy
.. automodule:: policy.ENACPolicy
.. automodule:: policy.TRACERPolicy
FeudalRL Policies
*********************
Traditional Reinforcement Learning algorithms fail to scale to large domains due to the curse of dimensionality. A novel Dialogue Management architecture based on Feudal RL decomposes the decision into two steps; a first step where a master policy selects a subset of primitive actions, and a second step where a primitive action is chosen from the selected subset. The structural information included in the domain ontology is used to abstract the dialogue state space, taking the decisions at each step using different
parts of the abstracted state. This, combined with an information sharing mechanism between slots, increases the scalability to large
domains.
For more information, please look at the paper `Feudal Reinforcement Learning for Dialogue Management in Large Domains <https://arxiv.org/pdf/1803.03232.pdf>`_.
\ No newline at end of file
Semantic Belief Tracking
************************
.. |.| raw:: html
<br />
.. automodule:: semanticbelieftracking.SemanticBeliefTrackingManager
:members:
:private-members:
.. automodule:: semanticbelieftracking.ModularSemanticBeliefTracker
.. autoclass:: semanticbelieftracking.ModularSemanticBeliefTracker.ModularSemanticBeliefTracker
Semantic Parsing
*********************
.. |.| raw:: html
<br />
.. automodule:: semi.SemI
:members:
:private-members:
.. automodule:: semi.RuleSemIMethods
.. autoclass:: semi.RuleSemIMethods.PassthroughSemI
.. autoclass:: semi.RuleSemIMethods.RegexSemI
.. automodule:: semi.RegexSemI
.. autoclass:: semi.RegexSemI.RegexSemI
Simulation
***************
.. |.| raw:: html
<br />
.. automodule:: usersimulator.SimulatedUsersManager
:members:
.. automodule:: usersimulator.ConfidenceScorer
:members:
.. automodule:: usersimulator.ConfusionModel
:members:
.. automodule:: usersimulator.ErrorModel
:members:
.. automodule:: usersimulator.NBestGenerator
:members:
.. automodule:: usersimulator.UserModel
:members:
.. automodule:: usersimulator.UMHdcSim
:members:
Dialogue Systems
*********************
.. |.| raw:: html
<br />
.. automodule:: DialogueServer
:members:
.. automodule:: Simulate
:members:
.. automodule:: Texthub
:members:
Functional Tests
***************************
.. |.| raw:: html
<br />
.. include:: ../tests/README.rst
List of Current Tests
======================
.. automodule:: tests.test_DialogueServer
:members:
.. automodule:: tests.test_Simulate
:members:
.. automodule:: tests.test_Tasks
:members:
\ No newline at end of file
Topic Tracking
***************
.. |.| raw:: html
<br />
.. automodule:: topictracking.TopicTracking
:members:
:private-members:
.. automodule:: topictracking.RuleTopicTrackers
:members:
Utility Modules
***************
.. |.| raw:: html
<br />
.. automodule:: utils.ContextLogger
:members:
.. automodule:: utils.dact
:members:
.. automodule:: utils.DiaAct
:members:
.. automodule:: utils.DialogueState
:members:
.. automodule:: utils.dummyDialogueServerClient
:members:
.. automodule:: utils.Scanner
:members:
.. automodule:: utils.Settings
:members:
.. PyDial documentation master file, created by
sphinx-quickstart on Mon Aug 10 14:07:56 2015.
You can adapt this file completely to your liking, but it should at least
contain the root `toctree` directive.
PyDial: the CUED Python Statistical Dialog System
*************************************************
Tutorials
====================
To get familiar with usage of PyDial, you can use the following tutorials:
- `Introduction to PyDial <tutorials/Introduction%20to%20PyDial.html>`_
- `Policy module <tutorials/Policy%20module.html>`_
- `Spoken Language Understanding <tutorials/Spoken%20Language%20Understanding.html>`_
- `Natural Language Generation <tutorials/Natural%20Language%20Generation%20Module.html>`_
- `How to add your own module <tutorials/How%20to%20add%20your%20own%20module.html>`_
- `How to add a new domain <tutorials/How%20to%20add%20a%20new%20domain.html>`_
- `Deep Reinforcement Learning Policies <tutorials/Deep%20RL%20Policies.html>`_
System Documentation
====================
.. toctree::
:maxdepth: 2
Agent
SemanticBeliefTracking
SemanticParsing
BeliefTracking
Policy
NLGeneration
Evaluation
TopicTracking
--------------------
.. toctree::
:maxdepth: 1
CEDM
--------------------
.. toctree::
:maxdepth: 2
Systems
Simulation
--------------------
.. toctree::
:maxdepth: 2
Ontology
Tests
Utilities
--------------------
.. toctree::
:maxdepth: 1
Acknowledgements
Indices and tables
==================
* :ref:`genindex`
* :ref:`modindex`
* :ref:`search`
#!/bin/bash
cp -r ../_build /tools/WorldWideWeb/old/research/dialogue/LocalDocs/cued-python-doc/
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment