hezar.models.image2text.vit_roberta package

Submodules

• hezar.models.image2text.vit_roberta.vit_roberta_image2text module
  • ViTRobertaImage2Text
    • ViTRobertaImage2Text.compute_loss()
    • ViTRobertaImage2Text.decoder
    • ViTRobertaImage2Text.encoder
    • ViTRobertaImage2Text.forward()
    • ViTRobertaImage2Text.generate()
    • ViTRobertaImage2Text.image_processor
    • ViTRobertaImage2Text.is_generative
    • ViTRobertaImage2Text.loss_func_name
    • ViTRobertaImage2Text.post_process()
    • ViTRobertaImage2Text.preprocess()
    • ViTRobertaImage2Text.required_backends
• hezar.models.image2text.vit_roberta.vit_roberta_image2text_config module
  • DecoderConfig
    • DecoderConfig.add_cross_attention
    • DecoderConfig.attention_probs_dropout_prob
    • DecoderConfig.bos_token_id
    • DecoderConfig.classifier_dropout
    • DecoderConfig.eos_token_id
    • DecoderConfig.gradient_checkpointing
    • DecoderConfig.hidden_act
    • DecoderConfig.hidden_dropout_prob
    • DecoderConfig.hidden_size
    • DecoderConfig.initializer_range
    • DecoderConfig.intermediate_size
    • DecoderConfig.is_decoder
    • DecoderConfig.layer_norm_eps
    • DecoderConfig.max_position_embeddings
    • DecoderConfig.num_attention_heads
    • DecoderConfig.num_hidden_layers
    • DecoderConfig.pad_token_id
    • DecoderConfig.position_embedding_type
    • DecoderConfig.type_vocab_size
    • DecoderConfig.use_cache
    • DecoderConfig.vocab_size
  • EncoderConfig
    • EncoderConfig.attention_probs_dropout_prob
    • EncoderConfig.encoder_stride
    • EncoderConfig.hidden_act
    • EncoderConfig.hidden_dropout_prob
    • EncoderConfig.hidden_size
    • EncoderConfig.image_size
    • EncoderConfig.initializer_range
    • EncoderConfig.intermediate_size
    • EncoderConfig.layer_norm_eps
    • EncoderConfig.num_attention_heads
    • EncoderConfig.num_channels
    • EncoderConfig.num_hidden_layers
    • EncoderConfig.patch_size
    • EncoderConfig.qkv_bias
  • GenerationConfig
    • GenerationConfig.bos_token_id
    • GenerationConfig.decoder_start_token_id
    • GenerationConfig.early_stopping
    • GenerationConfig.eos_token_id
    • GenerationConfig.length_penalty
    • GenerationConfig.max_length
    • GenerationConfig.no_repeat_ngram_size
    • GenerationConfig.num_beams
    • GenerationConfig.pad_token_id
    • GenerationConfig.return_dict_in_generate
  • ViTRobertaImage2TextConfig
    • ViTRobertaImage2TextConfig.decoder
    • ViTRobertaImage2TextConfig.encoder
    • ViTRobertaImage2TextConfig.generation
    • ViTRobertaImage2TextConfig.name

Module contents