hezar.models.image2text.vit_gpt2 package¶
Submodules¶
- hezar.models.image2text.vit_gpt2.vit_gpt2_image2text module
ViTGPT2Image2TextViTGPT2Image2Text.compute_loss()ViTGPT2Image2Text.forward()ViTGPT2Image2Text.generate()ViTGPT2Image2Text.image_processorViTGPT2Image2Text.is_generativeViTGPT2Image2Text.loss_func_nameViTGPT2Image2Text.post_process()ViTGPT2Image2Text.preprocess()ViTGPT2Image2Text.required_backendsViTGPT2Image2Text.tokenizer_name
- hezar.models.image2text.vit_gpt2.vit_gpt2_image2text_config module
DecoderConfigDecoderConfig.add_cross_attentionDecoderConfig.attn_pdropDecoderConfig.bos_token_idDecoderConfig.embd_pdropDecoderConfig.eos_token_idDecoderConfig.gradient_checkpointingDecoderConfig.initializer_rangeDecoderConfig.layer_norm_epsilonDecoderConfig.model_typeDecoderConfig.n_ctxDecoderConfig.n_embdDecoderConfig.n_headDecoderConfig.n_innerDecoderConfig.n_layerDecoderConfig.n_positionsDecoderConfig.nameDecoderConfig.resid_pdropDecoderConfig.summary_activationDecoderConfig.summary_first_dropoutDecoderConfig.use_cacheDecoderConfig.vocab_size
EncoderConfigEncoderConfig.attention_probs_dropout_probEncoderConfig.encoder_strideEncoderConfig.hidden_actEncoderConfig.hidden_dropout_probEncoderConfig.hidden_sizeEncoderConfig.image_sizeEncoderConfig.initializer_rangeEncoderConfig.intermediate_sizeEncoderConfig.layer_norm_epsEncoderConfig.nameEncoderConfig.num_attention_headsEncoderConfig.num_channelsEncoderConfig.num_hidden_layersEncoderConfig.patch_sizeEncoderConfig.qkv_bias
GenerationConfigViTGPT2Image2TextConfig