MCPcopy Create free account

hub / github.com/apple/ml-streambridge / functions

Functions393 in github.com/apple/ml-streambridge

Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:150
Method__init__
(self, config)
streambridge/model/multimodal_encoder/siglip_encoder.py:182
Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:261
Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:334
Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:410
Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:463
Method__init__
(self, config: SigLipVisionConfig)
streambridge/model/multimodal_encoder/siglip_encoder.py:489
Method__init__
(self, vision_tower, vision_tower_cfg, delay_load=False)
streambridge/model/multimodal_encoder/siglip_encoder.py:540
Method__init__
( self, dtype=torch.bfloat16, img_size=384, pooling_factor=4, ckpt = N
eval/streaming_models/online_oryx.py:44
Method__init__
( self, dtype=torch.bfloat16, img_size=384, pooling_factor=4, ckpt = N
eval/streaming_models/online_llava_ov.py:47
Method__init__
( self, dtype=torch.bfloat16, img_size=384, pooling_factor=2, ckpt = N
eval/streaming_models/online_qwen2_vl.py:46
Function_convert_to_rgb
(image)
streambridge/utils.py:163
Function_convert_to_rgb
(image)
eval/utils.py:15
Method_init_weights
Initialize the weights
streambridge/model/multimodal_resampler/qformer.py:675
Method_init_weights
(self, module)
streambridge/model/multimodal_encoder/qwen2_vit.py:1010
Method_init_weights
Initialize the weights
streambridge/model/multimodal_encoder/siglip_encoder.py:319
Method_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
streambridge/model/multimodal_resampler/qformer.py:716
Method_reorder_cache
(self, past, beam_idx)
streambridge/model/multimodal_resampler/qformer.py:1131
Methodappend_message
(self, role, message)
streambridge/conversation.py:187
Functionbuild_logger
(logger_name, logger_filename)
streambridge/utils.py:41
Methodconfig
(self)
streambridge/model/multimodal_resampler/masked_drop.py:51
Methodconfig
(self)
streambridge/model/multimodal_resampler/qformer.py:1286
Methodconfig
(self)
streambridge/model/multimodal_resampler/builder.py:21
Methodconfig
(self)
streambridge/model/multimodal_resampler/vlm_attention.py:334
Methodconfig
(self)
streambridge/model/multimodal_resampler/spatial_pool.py:40
Methodconfig
(self)
streambridge/model/multimodal_resampler/perceiver.py:88
Methodconfig
(self)
streambridge/model/multimodal_projector/builder.py:19
Methodconfig
(self)
streambridge/model/multimodal_encoder/oryx_vit.py:831
Methodcreate_custom_forward
(module)
streambridge/model/multimodal_resampler/qformer.py:544
Methodcustom_forward
(*inputs)
streambridge/model/multimodal_resampler/qformer.py:545
Methoddevice
(self)
streambridge/model/multimodal_encoder/oryx_vit.py:823
Methoddevice
(self)
streambridge/model/multimodal_encoder/siglip_encoder.py:606
Methoddict
(self)
streambridge/conversation.py:278
Functiondisable_torch_init
Disable the redundant torch default initialization to accelerate model creation.
streambridge/utils.py:117
Functiondisabled_train
Overwrite model.train with this function to make sure train/eval mode does not change anymore.
streambridge/model/multimodal_resampler/qformer.py:56
Functiondist_barrier
()
streambridge/utils.py:36
Methoddtype
(self)
streambridge/model/multimodal_encoder/oryx_vit.py:819
Methoddtype
(self)
streambridge/model/multimodal_encoder/siglip_encoder.py:601
Methoddummy_feature
(self)
streambridge/model/multimodal_encoder/oryx_vit.py:815
Methoddummy_feature
(self)
streambridge/model/multimodal_encoder/siglip_encoder.py:597
Functionevaluate
(rank, stream_data_split)
eval/eval_benchmarks/parallel_videomme.py:19
Functionevaluate
(rank, stream_data_split)
eval/eval_benchmarks/parallel_ovo_bench.py:18
Methodexpand2square
(pil_img, background_color=(122, 116, 104))
streambridge/conversation.py:193
Methodextra_repr
(self)
streambridge/model/multimodal_encoder/qwen2_vit.py:498
Methodfeed_forward_chunk
(self, attention_output)
streambridge/model/multimodal_resampler/qformer.py:487
Methodfeed_forward_chunk_query
(self, attention_output)
streambridge/model/multimodal_resampler/qformer.py:492
Methodflush
(self)
streambridge/utils.py:111
Methodforward
(self, image_features)
streambridge/model/activate_videollm.py:36
Methodforward
(self, samples)
streambridge/model/activate_videollm.py:194
Methodforward
(self, image_features, *args, **kwargs)
streambridge/model/multimodal_resampler/masked_drop.py:21
Methodforward
( self, input_ids=None, position_ids=None, query_embeds=None, past_key
streambridge/model/multimodal_resampler/qformer.py:89
Methodforward
( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi
streambridge/model/multimodal_resampler/qformer.py:180
Methodforward
(self, hidden_states, input_tensor)
streambridge/model/multimodal_resampler/qformer.py:296
Methodforward
( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi
streambridge/model/multimodal_resampler/qformer.py:333
Methodforward
(self, hidden_states)
streambridge/model/multimodal_resampler/qformer.py:369
Methodforward
(self, hidden_states, input_tensor)
streambridge/model/multimodal_resampler/qformer.py:382
Methodforward
( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi
streambridge/model/multimodal_resampler/qformer.py:413
Methodforward
( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi
streambridge/model/multimodal_resampler/qformer.py:506
Methodforward
(self, hidden_states)
streambridge/model/multimodal_resampler/qformer.py:609
Methodforward
(self, hidden_states)
streambridge/model/multimodal_resampler/qformer.py:628
Methodforward
(self, hidden_states)
streambridge/model/multimodal_resampler/qformer.py:649
Methodforward
(self, sequence_output)
streambridge/model/multimodal_resampler/qformer.py:660
Methodforward
r""" encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`): Seq
streambridge/model/multimodal_resampler/qformer.py:815
Methodforward
r""" encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`): Seq
streambridge/model/multimodal_resampler/qformer.py:998
Methodforward
r""" labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`): Labels for computing the masked l
streambridge/model/multimodal_resampler/qformer.py:1161
Methodforward
(self, image_features, *args, **kwargs)
streambridge/model/multimodal_resampler/qformer.py:1267
Methodforward
(self, x, *args, **kwargs)
streambridge/model/multimodal_resampler/builder.py:17
Methodforward
(self, image_features, prompts=None, image_counts=None, long_video=False)
streambridge/model/multimodal_resampler/vlm_attention.py:136
Methodforward
(self, image_features, images, *args, **kwargs)
streambridge/model/multimodal_resampler/spatial_pool.py:28
Methodforward
(self, image_features, forward_type, image_size=None, pooling_factor=None)
streambridge/model/multimodal_resampler/perceiver.py:35
Methodforward
( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N
streambridge/model/language_model/qwen2vl_qwen.py:46
Methodforward
( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N
streambridge/model/language_model/llava_qwen.py:68
Methodforward
( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N
streambridge/model/language_model/oryx_qwen.py:47
Methodforward
(self, x, *args, **kwargs)
streambridge/model/multimodal_projector/builder.py:15
Methodforward
(self, x)
streambridge/model/multimodal_projector/builder.py:33
Methodforward
(self, x, size=(16,16), x2=None, size2=(16, 16), modalities='image')
streambridge/model/multimodal_projector/builder.py:87
Methodforward
(self, x: torch.Tensor, cu_slens=None)
streambridge/model/multimodal_encoder/oryx_vit.py:146
Methodforward
(self, x: torch.Tensor)
streambridge/model/multimodal_encoder/oryx_vit.py:200
Methodforward
(self, x: torch.Tensor, cu_slens=None)
streambridge/model/multimodal_encoder/oryx_vit.py:248
Methodforward
(self, x, cal_attn_pool=False)
streambridge/model/multimodal_encoder/oryx_vit.py:621
Methodforward
(self, images, cal_attn_pool=False)
streambridge/model/multimodal_encoder/oryx_vit.py:809
Methodforward
(self, x, position_ids)
streambridge/model/multimodal_encoder/qwen2_vit.py:176
Methodforward
(self, seqlen: int)
streambridge/model/multimodal_encoder/qwen2_vit.py:280
Methodforward
(self, hidden_states: torch.Tensor)
streambridge/model/multimodal_encoder/qwen2_vit.py:303
Methodforward
(self, x: torch.Tensor)
streambridge/model/multimodal_encoder/qwen2_vit.py:328
Methodforward
(self, x)
streambridge/model/multimodal_encoder/qwen2_vit.py:343
Methodforward
( self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: torch.Tensor = None
streambridge/model/multimodal_encoder/qwen2_vit.py:355
Methodforward
( self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: torch.Tensor = None
streambridge/model/multimodal_encoder/qwen2_vit.py:389
Methodforward
( self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: torch.Tensor = None
streambridge/model/multimodal_encoder/qwen2_vit.py:412
Methodforward
(self, hidden_states, cu_seqlens, rotary_pos_emb)
streambridge/model/multimodal_encoder/qwen2_vit.py:473
Methodforward
(self, hidden_states)
streambridge/model/multimodal_encoder/qwen2_vit.py:491
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
streambridge/model/multimodal_encoder/qwen2_vit.py:574
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
streambridge/model/multimodal_encoder/qwen2_vit.py:671
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
streambridge/model/multimodal_encoder/qwen2_vit.py:793
Methodforward
Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (
streambridge/model/multimodal_encoder/qwen2_vit.py:910
Methodforward
(self, hidden_states: torch.Tensor, grid_thw: torch.Tensor)
streambridge/model/multimodal_encoder/qwen2_vit.py:1088
Methodforward
(self, pixel_values: torch.FloatTensor)
streambridge/model/multimodal_encoder/siglip_encoder.py:170
Methodforward
Input shape: Batch x Time x Channel
streambridge/model/multimodal_encoder/siglip_encoder.py:198
Methodforward
(self, hidden_states: torch.Tensor)
streambridge/model/multimodal_encoder/siglip_encoder.py:252
Methodforward
Args: hidden_states (`torch.FloatTensor`): Input to the layer of shape `(batch, seq_len, embed_dim)`.
streambridge/model/multimodal_encoder/siglip_encoder.py:270
← previousnext →201–300 of 393, ranked by callers