MCPcopy Create free account

hub / github.com/bytedance/GR-MG / functions

Functions177 in github.com/bytedance/GR-MG

Method__init__
( self, state_dim, act_dim, act_len, act_latent_di
policy/model/model.py:34
Method__init__
(self, drop_prob=None)
policy/model/vision_transformer.py:130
Method__init__
(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.)
policy/model/vision_transformer.py:158
Method__init__
(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0., dr
policy/model/vision_transformer.py:185
Method__init__
(self, img_size=224, patch_size=16, in_chans=3, embed_dim=768)
policy/model/vision_transformer.py:208
Method__init__
(self, img_size=[224], patch_size=16, in_chans=3, num_classes=0, embed_dim=768, depth=12, num
policy/model/vision_transformer.py:225
Method__init__
(self, in_dim, out_dim, use_bn=False, norm_last_layer=True, nlayers=3, hidden_dim=2048, bottleneck_dim=256)
policy/model/vision_transformer.py:347
Method__init__
(self, pad)
policy/data/ego4d_dataset.py:60
Method__init__
Constructor. Args: data_dir: root directory of the data preprocess: image preprocess function seq_len: seq
policy/data/ego4d_dataset.py:85
Method__init__
(self, pad)
policy/data/calvin_dataset.py:56
Method__init__
(self, data_dir, seq_len=10, act_len=5, f
policy/data/calvin_dataset.py:81
Method__init__
(self, configs)
policy/training/trainer.py:57
Method__init__
(self, ckpt_path, configs, device)
evaluate/wrapper/model_wrapper.py:30
Method__init__
Environment wrapper which returns raw observations. Args: abs_datasets_dir: absolute datset directory observation_sap
evaluate/wrapper/calvin_env_wrapper_raw.py:34
Method__init__
(self, ckpt_path, res=256)
goal_gen/evaluate.py:26
Method__init__
(self, now, logdir, ckptdir)
goal_gen/utils/utils.py:21
Method__init__
( self, vae, text_encoder, tokenizer, unet, scheduler,
goal_gen/utils/pipeline.py:40
Method__init__
(self, pretrained_model_dir, device, seed=123,
goal_gen/model/model.py:25
Method__init__
(self, data_dir, resolution=256, resolution_before_crop=288
goal_gen/data/calvindataset.py:28
Method__init__
(self, configs)
goal_gen/training/trainer.py:21
Method__len__
(self)
policy/data/ego4d_dataset.py:145
Method__len__
(self)
policy/data/calvin_dataset.py:220
Method__len__
(self)
goal_gen/data/calvindataset.py:87
Method_init_weights
Initialize the weights.
policy/model/gpt2.py:456
Method_init_weights
(self, m)
policy/model/vision_transformer.py:254
Method_init_weights
(self, m)
policy/model/vision_transformer.py:370
Method_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
policy/model/gpt2.py:740
Method_set_gradient_checkpointing
(self, module, value=False)
policy/model/gpt2.py:483
Functionadjust_learning_rate
Decay the learning rate with half-cycle cosine after warmup
policy/training/trainer.py:25
Methodconfigure_optimizers
(self)
policy/training/trainer.py:180
Methodconfigure_optimizers
(self)
goal_gen/training/trainer.py:68
Methodcreate_custom_forward
(module)
policy/model/gpt2.py:881
Methodcustom_forward
(*inputs)
policy/model/gpt2.py:882
Methoddeparallelize
(self)
policy/model/gpt2.py:718
Methodforward
( self, hidden_states: Optional[Tuple[torch.FloatTensor]], layer_past: Optional[Tuple[
policy/model/gpt2.py:290
Methodforward
(self, hidden_states: Optional[Tuple[torch.FloatTensor]])
policy/model/gpt2.py:353
Methodforward
( self, hidden_states: Optional[Tuple[torch.FloatTensor]], layer_past: Optional[Tuple[
policy/model/gpt2.py:377
Methodforward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Optional[Tuple[
policy/model/gpt2.py:753
Methodforward
(self, x)
policy/model/vision_transformer.py:134
Methodforward
(self, x)
policy/model/vision_transformer.py:148
Methodforward
(self, x)
policy/model/vision_transformer.py:169
Methodforward
(self, x, return_attention=False)
policy/model/vision_transformer.py:196
Methodforward
(self, x)
policy/model/vision_transformer.py:217
Methodforward
(self, x)
policy/model/vision_transformer.py:298
Methodforward
(self, x)
policy/model/vision_transformer.py:376
Methodforward
(self, x)
policy/data/ego4d_dataset.py:41
Methodforward
(self, x)
policy/data/ego4d_dataset.py:64
Methodforward
(self, x)
policy/data/calvin_dataset.py:36
Methodforward
(self, x)
policy/data/calvin_dataset.py:60
Methodfrom_checkpoint
(cls, ckpt_dir=None, configs=None)
goal_gen/training/trainer.py:63
Methodget_input_embeddings
(self)
policy/model/gpt2.py:734
Methodget_intermediate_layers
(self, x, n=1)
policy/model/vision_transformer.py:314
Methodget_last_selfattention
(self, x)
policy/model/vision_transformer.py:305
Functionload_tf_weights_in_gpt2
Load tf checkpoints in a pytorch model
policy/model/gpt2.py:67
Methodnum_gpus
(self)
policy/training/trainer.py:72
Methodnum_gpus
(self)
goal_gen/training/trainer.py:36
Methodon_load_checkpoint
(self, checkpoint)
goal_gen/training/trainer.py:123
Methodon_save_checkpoint
(self, checkpoint)
goal_gen/training/trainer.py:118
Methodon_train_start
(self, trainer, model)
policy/main.py:52
Methodon_train_start
(self, trainer, model)
goal_gen/utils/utils.py:27
Methodparallelize
(self, device_map=None)
policy/model/gpt2.py:691
Functionreduce_mean
(tensor, nprocs=None)
policy/utils/dist_train.py:51
Functionreduce_mean
(tensor, nprocs=None)
goal_gen/utils/dist_train.py:51
Functionreduce_sum
(tensor)
policy/utils/dist_train.py:66
Functionreduce_sum
(tensor)
goal_gen/utils/dist_train.py:66
Methodreset
( self, reset_info: Dict[str, Any] = None, batch_idx: int = 0, seq_idx: int =
evaluate/wrapper/calvin_env_wrapper_raw.py:90
Functionretrieve_latents
( encoder_output: torch.Tensor, generator: Optional[torch.Generator] = None, sample_mode: str = "sample" )
goal_gen/utils/pipeline.py:23
Methodset_egl_device
(device)
evaluate/wrapper/calvin_env_wrapper_raw.py:52
Methodset_input_embeddings
(self, new_embeddings)
policy/model/gpt2.py:737
Methodstep
( self, action_tensor: torch.Tensor )
evaluate/wrapper/calvin_env_wrapper_raw.py:68
Methodtraining_step
(self, batch, batch_idx)
policy/training/trainer.py:364
Methodtraining_step
(self, batch, batch_idx)
goal_gen/training/trainer.py:108
Methodvalidation_step
(self, batch, batch_idx)
policy/training/trainer.py:223
Methodvalidation_step
(self, batch, batch_idx, dataloader_idx=0)
goal_gen/training/trainer.py:96
Functionvit_base
(patch_size=16, **kwargs)
policy/model/vision_transformer.py:339
Functionvit_small
(patch_size=16, **kwargs)
policy/model/vision_transformer.py:332
Functionvit_tiny
(patch_size=16, **kwargs)
policy/model/vision_transformer.py:325
← previous101–177 of 177, ranked by callers