Logo
1.0.0rc1
  • Installation
    • Docker Images
  • QuickStart
    • Command line interface (CLI)
    • Python
    • Using GPU
    • Pretrained models
    • Docker images
    • Out-of-the-box pretrained models
      • Text Question Answering
      • Name Entity Recognition
      • Insult Detection
      • Paraphrase Detection
  • General concepts
    • Key Concepts
  • Configuration file
    • Nested configuration files
    • Variables
    • Training
      • Train config
      • Train Parameters
        • Metrics
      • DatasetReader
      • DataLearningIterator and DataFittingIterator
    • Inference
    • Model Configuration
      • Preprocessors
      • Tokenizers
      • Embedders
      • Vectorizers
  • Models/Skills overview
    • Models
      • NER model [docs]
      • Classification model [docs]
      • Automatic spelling correction model [docs]
      • Ranking model [docs]
      • TF-IDF Ranker model [docs]
      • Question Answering model [docs]
      • Frequently Asked Questions (FAQ) model [docs]
    • Skills
      • ODQA [docs]
    • AutoML
      • Hyperparameters optimization [docs]
    • Embeddings
      • Pre-trained embeddings [docs]
    • Examples of some models

Features

  • Pre-trained embeddings
    • BERT
      • License
      • Downloads
    • ELMo
      • Downloads
    • fastText
      • License
      • Downloads
      • Word vectors training parameters
  • AutoML
    • Cross-validation
      • Parameters
      • Special parameters in config
      • Results

Models

  • BERT-based models
    • BERT as Embedder
    • BERT for Classification
    • BERT for Named Entity Recognition (Sequence Tagging)
    • BERT for Context Question Answering (SQuAD)
    • Using custom BERT in DeepPavlov
  • Context Question Answering
    • Task definition
    • Models
      • BERT
    • Configuration
    • Prerequisites
    • Model usage from Python
    • Model usage from CLI
      • Training
      • Interact mode
    • Pretrained models:
      • SQuAD
      • SQuAD with contexts without correct answers
      • SDSJ Task B
  • Classification
    • Quick start
      • Command line
      • Python code
    • BERT models
    • Neural Networks on PyTorch
    • Sklearn models
    • Pre-trained models
    • GLUE Benchmark
    • How to train on other datasets
    • How to improve the performance
    • References
  • Entity Extraction
    • Use the model
    • Use the model
  • Named Entity Recognition
    • Train and use the model
    • Multilingual BERT Zero-Shot Transfer
    • NER task
    • Training data
    • NER-based Model for Sentence Boundary Detection Task
    • Multilingual Case-insensitive Named Entity Recognition
    • Literature
  • Neural Ranking
    • Training and inference models on predifined datasets
      • BERT Ranking
      • Paraphrase identification
      • Paraphraser.ru dataset
      • Paraphrase identification
  • Spelling Correction
    • Quick start
    • levenshtein_corrector
      • Component config parameters:
    • brillmoore
      • Component config parameters:
      • Training configuration
    • Language model
    • Comparison
  • TF-IDF Ranking
    • Quick Start
    • Configuration
    • Running the Ranker
      • Training
      • Interacting
    • Available Data and Pretrained Models
      • enwiki.db
      • enwiki_tfidf_matrix.npz
      • ruwiki.db
      • ruwiki_tfidf_matrix.npz
    • Comparison
    • References
  • Popularity Ranking
    • Quick Start
    • Configuration
    • Running the Ranker
      • Interacting
    • Available Data and Pretrained Models
    • References
  • Knowledge Base Question answering
    • Overview
    • Built-In Models
    • How Do I: Using KBQA In CLI & Python
    • How Do I: Train KBQA Model
    • How Do I: Train Query Prediction Model
    • How Do I: Train Entity Detection Model
    • How Do I: Train Relation and Path Ranking Models
    • How Do I: Adding Templates For New SPARQL Queries
    • Advanced: Using Wiki Parser As Standalone Service For KBQA
  • Relation Extraction
    • English RE model
    • Russian RE model
    • RE Model Architecture
  • SuperGLUE Submission
    • Task definition
    • Train your model
    • Create your submission files
    • Scores

Skills

  • Open-Domain Question Answering
    • Task definition
    • Quick Start
    • Languages
    • Models
    • Running ODQA
      • Training
      • Interacting
    • Configuration
    • Comparison
    • References
  • Frequently Asked Questions Answering
    • Quick Start
      • Building
      • Inference
    • Config
      • Config Structure
      • Vectorizers
      • Classifiers for FAQ
    • Running FAQ
      • Training
      • Interacting
    • Available Data and Pretrained Models

Integrations

  • REST API
    • API routes
      • /model
      • /probe
      • /api
      • /docs
      • /metrics
    • Advanced configuration
  • Socket API
    • Advanced configuration
    • Socket client example (Python)
  • DeepPavlov Agent RabbitMQ integration
    • Command line interface
    • Python interface
  • Amazon AWS deployment
    • 1. AWS EC2 machine launch
    • 2. DeepPavlov ODQA deployment
    • 3. Accessing your ODQA API
  • DeepPavlov settings
    • 1. Settings files access and management
    • 2. Dialog logging
    • 3. Environment variables

Developer Guides

  • Contribution guide
  • Register your model

Internships

  • Internships

Package Reference

  • core
    • deeppavlov.core.commands
    • deeppavlov.core.common
    • deeppavlov.core.data
    • deeppavlov.core.models
    • deeppavlov.core.trainers
  • dataset_iterators
  • dataset_readers
  • metrics
  • models
    • deeppavlov.models.api_requester
    • deeppavlov.models.classifiers
    • deeppavlov.models.doc_retrieval
    • deeppavlov.models.embedders
    • deeppavlov.models.entity_extraction
    • deeppavlov.models.kbqa
    • deeppavlov.models.preprocessors
    • deeppavlov.models.relation_extraction
    • deeppavlov.models.sklearn
    • deeppavlov.models.spelling_correction
    • deeppavlov.models.tokenizers
    • deeppavlov.models.torch_bert
    • deeppavlov.models.vectorizers
  • vocabs
DeepPavlov
  • »
  • core
  • Edit on GitHub

coreΒΆ

DeepPavlov Core

Core

  • deeppavlov.core.commands
  • deeppavlov.core.common
  • deeppavlov.core.data
  • deeppavlov.core.models
  • deeppavlov.core.trainers
Next Previous

Problem? Ask a Question or try our Demo

twitter youtube medium

© Copyright 2018, Neural Networks and Deep Learning lab, MIPT Revision bfd5cd06.

Built with Sphinx using a theme provided by Read the Docs.