MCPcopy Create free account

hub / github.com/SJTU-DENG-Lab/WLA / functions

Functions234 in github.com/SJTU-DENG-Lab/WLA

⨍ Functions 234◇ Types & classes 41
Methodforward
( self, hidden_states: torch.Tensor, position_embeddings: Optional[Tuple[torch.Tensor,
models/transformer_encoder.py:100
Methodforward
(self, hidden_states)
models/transformer_encoder.py:150
Methodforward
(self, hidden_states, timestep, encoder_hidden_states=None, encoder_attention_mask=None)
models/model.py:527
Methodforward
(self, timesteps)
models/action_model/action_encoder.py:34
Methodforward
actions: shape (B, T, action_dim) timesteps: shape (B,) -- a single scalar per batch item returns: shape (B, T, hidden_s
models/action_model/action_encoder.py:69
Methodforward
(self, x, cat_ids)
models/action_model/action_model.py:33
Methodforward
(self, x, cat_ids)
models/action_model/action_model.py:47
Methodforward
(self, x)
models/action_model/action_model.py:59
Methodforward
actions: shape (B, T, action_dim) timesteps: shape (B,) -- a single scalar per batch item returns: shape (B, T, hidden_s
models/action_model/action_model.py:73
Methodforward
vl_embs: list of torch.Tensor, each shape (B, seq_length, feature_dim) actions: shape (B, future_action_window_size, D_action)
models/action_model/action_model.py:210
Methodforward
(self, timesteps)
models/action_model/cross_attention_dit.py:37
Methodforward
( self, x: torch.Tensor, temb: Optional[torch.Tensor] = None, )
models/action_model/cross_attention_dit.py:59
Methodforward
( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,
models/action_model/cross_attention_dit.py:150
Methodforward
( self, hidden_states: torch.Tensor, # Shape: (B, T, D) encoder_hidden_states: torch.
models/action_model/cross_attention_dit.py:266
Methodforward
( self, hidden_states: torch.Tensor, # Shape: (B, T, D) return_all_hidden_states: boo
models/action_model/cross_attention_dit.py:362
Functionfreeze_hook
(grad, mode)
train.py:217
Functionget_action_model
Factory: build FlowmatchingActionHead from global framework config. Args: config: Global config (expects config.framework.action
models/action_model/action_model.py:321
Functionget_model
(usr_args)
experiments/robotwin/deploy_policy.py:16
Functionget_model
(usr_args)
experiments/rmbench/deploy_policy.py:16
Methodget_output_embeddings
(self)
models/wla.py:110
Methodget_resize_fn
(self)
models/model.py:205
Methodget_tokenize_fn
(self)
models/model.py:202
Methodget_tokenizer
(self)
models/model.py:199
Methodget_vision_positions
(input_ids, token_id, image_indices)
models/model.py:473
Methodon_train_begin
(self, args, state, control, model, **kwargs)
trainer.py:389
Methodprediction_step
( self, model: nn.Module, inputs: Dict[str, Union[torch.Tensor, Any]], predict
trainer.py:179
Methodprepare_input
(self, batch: dict)
models/action_model/action_model.py:154
Functionreset_model
(model)
experiments/robotwin/deploy_policy.py:24
Functionreset_model
(model)
experiments/rmbench/deploy_policy.py:26
Functionseed_everything
(seed=42)
trainer_utils.py:57
Methodset_input_embeddings
(self, value)
models/wla.py:107
Methodset_output_embeddings
(self, new_embeddings)
models/wla.py:113
Methodtokenize
( tokenizer, caption, image=None, text_response=None, add_generation_p
models/model.py:210
Methodwla_eval_data_collator
(features)
trainer.py:116
← previous201–234 of 234, ranked by callers