OpenNMT-tf Logo
2.32

Getting started

  • Quickstart
  • Installation

Configuration

  • Model
  • Parameters
  • Data
  • Vocabulary
  • Tokenization
  • Embeddings
  • Alignments

Usage

  • Training
  • Inference
  • Serving
  • 2.0 Transition Guide
  • Frequently Asked Questions

API

  • Python
    • opennmt
    • opennmt.data
      • Noise
      • Vocab
      • WordDropout
      • WordNoiser
      • WordOmission
      • WordPermutation
      • WordReplacement
      • alignment_matrix_from_pharaoh
      • batch_dataset
      • batch_sequence_dataset
      • create_lookup_tables
      • filter_examples_by_length
      • filter_irregular_batches
      • get_dataset_size
      • inference_pipeline
      • make_cardinality_multiple_of
      • random_shard
      • shuffle_dataset
      • tokens_to_chars
      • tokens_to_words
      • training_pipeline
    • opennmt.decoders
    • opennmt.encoders
    • opennmt.inputters
    • opennmt.layers
    • opennmt.models
    • opennmt.optimizers
    • opennmt.schedules
    • opennmt.tokenizers
    • opennmt.utils
OpenNMT-tf
  • »
  • Python »
  • opennmt.data

opennmt.data

This module exposes classes and functions that help creating and processing data.

  • Noise
  • Vocab
  • WordDropout
  • WordNoiser
  • WordOmission
  • WordPermutation
  • WordReplacement
  • alignment_matrix_from_pharaoh
  • batch_dataset
  • batch_sequence_dataset
  • create_lookup_tables
  • filter_examples_by_length
  • filter_irregular_batches
  • get_dataset_size
  • inference_pipeline
  • make_cardinality_multiple_of
  • random_shard
  • shuffle_dataset
  • tokens_to_chars
  • tokens_to_words
  • training_pipeline
Previous Next