lmflow.models.vision2seq_model

Classes

CustomAutoVision2SeqModel

Helper class that provides a standard way to create an ABC using

Module Contents

class lmflow.models.vision2seq_model.CustomAutoVision2SeqModel(config: transformers.Blip2Config, image_encoder_name_or_path=None, qformer_name_or_path=None, language_model_name_or_path=None, low_resource=False)

Bases: transformers.Blip2ForConditionalGeneration, lmflow.models.base_model.BaseModel

Helper class that provides a standard way to create an ABC using inheritance.

custom_vision_model

with_qformer

language_model

hidden_size

get_backend_model()

vision_model_from_pretrained(pretrained_path)

qformer_from_pretrained(pretrained_path)

language_model_from_pretrained(pretrained_path, low_resource=False, use_prompt_cache=False)

vision_feature_select(image_forward_outs)

register_prompt_cache(prompt_ids, prompt_keys_values)

Udpate the prompt id and embedding for reuse in the future

Args:

prompt_ids (torch.LongTensor): The id of the prompt. prompt_keys_values (torch.FloatTensor): The embedding of the prompt.

Returns:

None

save_prompt_cache(path)

Save prompt embedding and id.

Args:

path: The path to save the prompt embedding and id.

Returns:

None

load_prompt_cache(path)

Load prompt embedding and id. Args:

path: The path to load the prompt embedding and id.

Returns:

None

get_tokenizer()

forward(input_ids: torch.LongTensor = None, pixel_values: torch.FloatTensor | None = None, images: torch.FloatTensor | None = None, attention_mask: torch.Tensor | None = None, past_key_values: List[torch.FloatTensor] | None = None, inputs_embeds: torch.FloatTensor | None = None, labels: torch.LongTensor | None = None, use_cache: bool | None = None, output_attentions: bool | None = None, output_hidden_states: bool | None = None, return_dict: bool | None = None, image_token_indexes: List | None = [0], one_sample_multiple_images: bool = False) → Tuple | transformers.modeling_outputs.CausalLMOutputWithPast

processor_image_token_in_minigpt4(input_ids, language_model_inputs, attention_mask, image_token_indexes, pixel_values, batch_size=1)

generate(pixel_values: torch.FloatTensor, input_ids: torch.LongTensor | None = None, attention_mask: torch.LongTensor | None = None, image_token_indexes: List | None = [0], one_sample_multiple_images: bool | None = False, images: torch.LongTensor | None = None, **generate_kwargs) → torch.LongTensor

Overrides generate function to be able to use the model as a conditional generator.

Args:

pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)):

Input images to be processed.

input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional):

The sequence used as a prompt for the generation.

attention_mask (torch.LongTensor of shape (batch_size, sequence_length), optional):

Mask to avoid performing attention on padding token indices

image_token_indexes (bool, optional):

The index for inserting the image tokens.

one_sample_multiple_images: (bool, optional):

The flag for inference that the input batch size is 1 and contain multiple images.

Returns:

captions (list): A list of strings of length batch_size * num_captions.