Choosing a Pipeline

These docs are for version 1.x of Rasa Open Source.

User Guide

• Installation
• Tutorial: Rasa Basics
• Tutorial: Building Assistants
• Command Line Interface
• Architecture
• Messaging and Voice Channels
• Testing Your Assistant
• Setting up CI/CD
• Validate Data
• Configuring the HTTP API
• Deploying Your Rasa Assistant
• Cloud Storage

NLU

• About
• Using NLU Only
• Training Data Format
• Language Support
• Choosing a Pipeline
• Components
• Entity Extraction

Core

• About
• Stories
• Domains
• Responses
• Actions
• Reminders and External Events
• Policies
• Slots
• Forms
• Retrieval Actions
• Interactive Learning
• Fallback Actions
• Knowledge Base Actions

Conversation Design

• Dialogue Elements
• Small Talk
• Completing Tasks
• Guiding Users

API Reference

• Action Server
• HTTP API
• Jupyter Notebooks
• Agent
• Custom NLU Components
• Rasa SDK
• Events
• Tracker
• Tracker Stores
• Event Brokers
• Lock Stores
• Training Data Importers
• Featurization of Conversations
• TensorFlow Configuration
• Migration Guide
• Rasa Open Source Change Log

Migrate from (beta)

• Dialogflow
• Wit.ai
• LUIS
• IBM Watson

Reference

• Glossary

Versions

viewing: 1.10.13

Choosing a Pipeline

In Rasa Open Source, incoming messages are processed by a sequence of components. These components are executed one after another in a so-called processing pipeline defined in your config.yml. Choosing an NLU pipeline allows you to customize your model and finetune it on your dataset.

How to Choose a Pipeline

If your training data is in English, a good starting point is the following pipeline:

language: "en"

pipeline:
  - name: ConveRTTokenizer
  - name: ConveRTFeaturizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 100
  - name: EntitySynonymMapper
  - name: ResponseSelector
    epochs: 100

If your training data is not in English, start with the following pipeline:

language: "fr"  # your two-letter language code

pipeline:
  - name: WhitespaceTokenizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 100
  - name: EntitySynonymMapper
  - name: ResponseSelector
    epochs: 100

Choosing the Right Components

There are components for entity extraction, for intent classification, response selection, pre-processing, and others. You can learn more about any specific component on the Components page. If you want to add your own component, for example to run a spell-check or to do sentiment analysis, check out Custom NLU Components.

A pipeline usually consists of three main parts:

• Tokenization

• Featurization

• Entity Recognition / Intent Classification / Response Selectors

Tokenization

For tokenization of English input, we recommend the ConveRTTokenizer. You can process other whitespace-tokenized (words are separated by spaces) languages with the WhitespaceTokenizer. If your language is not whitespace-tokenized, you should use a different tokenizer. We support a number of different tokenizers, or you can create your own custom tokenizer.

Featurization

You need to decide whether to use components that provide pre-trained word embeddings or not. We recommend in cases of small amounts of training data to start with pre-trained word embeddings. Once you have a larger amount of data and ensure that most relevant words will be in your data and therefore will have a word embedding, supervised embeddings, which learn word meanings directly from your training data, can make your model more specific to your domain. If you can’t find a pre-trained model for your language, you should use supervised embeddings.

• Pre-trained Embeddings

• Supervised Embeddings

Entity Recognition / Intent Classification / Response Selectors

Depending on your data you may want to only perform intent classification, entity recognition or response selection. Or you might want to combine multiple of those tasks. We support several components for each of the tasks. All of them are listed in Components. We recommend using DIETClassifier for intent classification and entity recognition and ResponseSelector for response selection.

Multi-Intent Classification

You can use Rasa Open Source components to split intents into multiple labels. For example, you can predict multiple intents (thank+goodbye) or model hierarchical intent structure (feedback+positive being more similar to feedback+negative than chitchat). To do this, you need to use the DIETClassifier in your pipeline. You’ll also need to define these flags in whichever tokenizer you are using:

• intent_tokenization_flag: Set it to True, so that intent labels are tokenized.

• intent_split_symbol: Set it to the delimiter string that splits the intent labels. In this case +, default _.

Comparing Pipelines

Rasa gives you the tools to compare the performance of multiple pipelines on your data directly. See Comparing NLU Pipelines for more information.

Handling Class Imbalance

Classification algorithms often do not perform well if there is a large class imbalance, for example if you have a lot of training data for some intents and very little training data for others. To mitigate this problem, you can use a balanced batching strategy. This algorithm ensures that all classes are represented in every batch, or at least in as many subsequent batches as possible, still mimicking the fact that some classes are more frequent than others. Balanced batching is used by default. In order to turn it off and use a classic batching strategy include batch_strategy: sequence in your config file.