Logo
latest
  • Overview
    • Key Features
    • Library API Example
    • Installation
    • Getting Started
    • Reference
    • License
  • Examples
    • Examples by Models/Algorithms
      • RNN / Seq2seq
      • Transformer (Self-attention)
      • Variational Autoencoder (VAE)
      • Classifier / Sequence Prediction
    • Examples by Tasks
      • Language Modeling
      • Machine Translation
      • Classification
  • HParams
  • Data
    • Tokenizer
      • TokenizerBase
      • SentencePieceTokenizer
      • BERTTokenizer
      • GPT2Tokenizer
      • RoBERTaTokenizer
      • XLNetTokenizer
      • T5Tokenizer
    • Vocabulary
      • SpecialTokens
      • Vocab
      • map_ids_to_strs
    • Embedding
      • Embedding
      • load_word2vec
      • load_glove
    • Data Sources
      • DataSource
      • SequenceDataSource
      • IterDataSource
      • ZipDataSource
      • FilterDataSource
      • RecordDataSource
      • TextLineDataSource
      • PickleDataSource
    • Data Loaders
      • DatasetBase
      • MonoTextData
      • PairedTextData
      • ScalarData
      • MultiAlignedData
      • RecordData
    • Data Iterators
      • Batch
      • DataIterator
      • TrainTestDataIterator
      • BatchingStrategy
      • TokenCountBatchingStrategy
    • Data Utilities
      • maybe_download
      • read_words
      • make_vocab
      • count_file_lines
  • Core
    • Attention Mechanism
      • AttentionWrapperState
      • LuongAttention
      • BahdanauAttention
      • BahdanauMonotonicAttention
      • LuongMonotonicAttention
      • compute_attention
      • monotonic_attention
      • hardmax
      • sparsemax
    • Cells
      • default_rnn_cell_hparams
      • get_rnn_cell
      • wrap_builtin_cell
      • RNNCellBase
      • RNNCell
      • GRUCell
      • LSTMCell
      • DropoutWrapper
      • ResidualWrapper
      • HighwayWrapper
      • MultiRNNCell
      • AttentionWrapper
    • Layers
      • get_layer
      • MaxReducePool1d
      • AvgReducePool1d
      • get_pooling_layer_hparams
      • MergeLayer
      • Flatten
      • Identity
      • default_regularizer_hparams
      • get_regularizer
      • get_initializer
      • get_activation_fn
    • Optimization
      • default_optimization_hparams
      • get_train_op
      • get_scheduler
      • get_optimizer
      • get_grad_clip_fn
  • Modules
    • ModuleBase
    • Embedders
      • WordEmbedder
      • PositionEmbedder
      • SinusoidsPositionEmbedder
      • EmbedderBase
    • Encoders
      • UnidirectionalRNNEncoder
      • BidirectionalRNNEncoder
      • MultiheadAttentionEncoder
      • TransformerEncoder
      • BERTEncoder
      • RoBERTaEncoder
      • GPT2Encoder
      • XLNetEncoder
      • Conv1DEncoder
      • EncoderBase
      • RNNEncoderBase
      • default_transformer_poswise_net_hparams
    • Decoders
      • DecoderBase
      • RNNDecoderBase
      • BasicRNNDecoder
      • BasicRNNDecoderOutput
      • AttentionRNNDecoder
      • AttentionRNNDecoderOutput
      • GPT2Decoder
      • XLNetDecoder
      • XLNetDecoderOutput
      • TransformerDecoder
      • TransformerDecoderOutput
      • Helper
      • TrainingHelper
      • EmbeddingHelper
      • GreedyEmbeddingHelper
      • SampleEmbeddingHelper
      • TopKSampleEmbeddingHelper
      • TopPSampleEmbeddingHelper
      • SoftmaxEmbeddingHelper
      • GumbelSoftmaxEmbeddingHelper
      • get_helper
    • Classifiers
      • BERTClassifier
      • RoBERTaClassifier
      • GPT2Classifier
      • UnidirectionalRNNClassifier
      • Conv1DClassifier
      • XLNetClassifier
    • Regressors
      • XLNetRegressor
    • EncoderDecoders
      • T5EncoderDecoder
    • Pre-trained
      • PretrainedMixin
      • PretrainedBERTMixin
      • PretrainedRoBERTaMixin
      • PretrainedGPT2Mixin
      • PretrainedXLNetMixin
      • PretrainedT5Mixin
    • Connectors
      • ConnectorBase
      • ConstantConnector
      • ForwardConnector
      • MLPTransformConnector
    • Networks
      • FeedForwardNetworkBase
      • FeedForwardNetwork
      • Conv1DNetwork
  • Loss Functions
    • MLE Loss
      • sequence_softmax_cross_entropy
      • sequence_sparse_softmax_cross_entropy
      • sequence_sigmoid_cross_entropy
      • binary_sigmoid_cross_entropy
      • binary_sigmoid_cross_entropy_with_clas
    • Policy Gradient Loss
      • pg_loss_with_logits
      • pg_loss_with_log_probs
    • Reward
      • discount_reward
    • Adversarial Loss
      • binary_adversarial_losses
    • Entropy
      • entropy_with_logits
      • sequence_entropy_with_logits
    • Loss Utilities
      • mask_and_reduce
      • reduce_batch_time
      • reduce_dimensions
  • Executor
    • Executor
      • Executor
    • Conditions
      • Event
      • Condition
      • epoch
      • iteration
      • validation
      • consecutive
      • once
      • time
    • Metrics
      • Metric
      • SimpleMetric
      • StreamingMetric
      • Accuracy
      • ConfusionMatrix
      • Precision
      • Recall
      • F1
      • PearsonR
      • RMSE
      • Average
      • AveragePerplexity
      • RunningAverage
      • LR
    • Actions
      • reset_params
      • scale_lr
      • early_stop
  • Evaluations
    • BLEU
      • sentence_bleu
      • corpus_bleu
      • sentence_bleu_moses
      • corpus_bleu_moses
      • corpus_bleu_transformer
      • bleu_transformer_tokenize
      • file_bleu
    • Accuracy
      • accuracy
      • binary_clas_accurac
  • Utilities
    • Frequent Use
      • AverageRecorder
      • compat_as_text
      • write_paired_text
    • IO
      • write_paired_text
      • maybe_create_dir
    • DType
      • get_numpy_dtype
      • maybe_hparams_to_dict
      • compat_as_text
    • Shape
      • mask_sequences
      • transpose_batch_time
      • flatten
      • pad_and_concat
    • Dictionary
      • dict_patch
      • dict_lookup
      • dict_fetch
      • dict_pop
      • flatten_dict
    • String
      • strip_token
      • strip_eos
      • strip_special_tokens
      • str_join
      • default_str
      • uniquify_str
    • Meta
      • check_or_get_class
      • get_class
      • check_or_get_instance
      • get_instance
      • check_or_get_instance_with_redundant_kwargs
      • get_instance_with_redundant_kwargs
      • get_function
      • call_function_with_redundant_kwargs
      • get_args
      • get_default_arg_values
      • get_instance_kwargs
    • Misc
      • ceildiv
      • map_structure
      • map_structure_zip
      • beam_search
      • flatten
      • pack_sequence_as
    • AverageRecorder
Texar
  • Docs »
  • Welcome to Texar-PyTorch’s documentation!
  • Edit on GitHub

Welcome to Texar-PyTorch’s documentation!¶

Texar is a modularized, versatile, and extensible toolkit for machine learning and text generation tasks.

If you work with TensorFlow, check out the documentation of Texar (TensorFlow).

  • Overview
  • Examples
    • Examples by Models/Algorithms
    • Examples by Tasks

API¶

  • HParams
  • Data
    • Tokenizer
    • Vocabulary
    • Embedding
    • Data Sources
    • Data Loaders
    • Data Iterators
    • Data Utilities
  • Core
    • Attention Mechanism
    • Cells
    • Layers
    • Optimization
  • Modules
    • ModuleBase
    • Embedders
    • Encoders
    • Decoders
    • Classifiers
    • Regressors
    • EncoderDecoders
    • Pre-trained
    • Connectors
    • Networks
  • Loss Functions
    • MLE Loss
    • Policy Gradient Loss
    • Reward
    • Adversarial Loss
    • Entropy
    • Loss Utilities
  • Executor
    • Executor
    • Conditions
    • Metrics
    • Actions
  • Evaluations
    • BLEU
    • Accuracy
  • Utilities
    • Frequent Use
    • IO
    • DType
    • Shape
    • Dictionary
    • String
    • Meta
    • Misc
    • AverageRecorder
Next

© Copyright 2019, Texar Revision 5b56ad39.

Built with Sphinx using a theme provided by Read the Docs.
Read the Docs v: latest
Versions
latest
stable
Downloads
On Read the Docs
Project Home
Builds

Free document hosting provided by Read the Docs.