MCPcopy Create free account

hub / github.com/cientgu/VQ-Diffusion / functions

Functions543 in github.com/cientgu/VQ-Diffusion

⨍ Functions 543◇ Types & classes 107
Functionlogits_top_k
(logits, filter_ratio = 0.5, minimum=1, pad_value=None)
image_synthesis/modeling/utils/misc.py:9
Functionmain_worker
(local_rank, args)
train.py:128
Functionmask_with_top_k
mask the input tensor along the last dimension. The values the not in the topk will be masked as zeros
image_synthesis/modeling/utils/misc.py:27
Methodparameters
Following minGPT: This long function is unfortunately doing something very simple and is being very defensive: We are separat
image_synthesis/modeling/transformers/diffusion_transformer.py:367
Methodparameters
Following minGPT: This long function is unfortunately doing something very simple and is being very defensive: We are separat
image_synthesis/modeling/transformers/diffusion_transformer_amp.py:372
Methodparameters
Following minGPT: This long function is unfortunately doing something very simple and is being very defensive: We are sepa
image_synthesis/modeling/transformers/transformer_utils.py:510
Methodparameters
Following minGPT: This long function is unfortunately doing something very simple and is being very defensive: We are sepa
image_synthesis/modeling/transformers/transformer_utils.py:650
Methodparameters
(self, recurse=True, name=None)
image_synthesis/modeling/models/unconditional_dalle.py:33
Methodparameters
(self, recurse=True, name=None)
image_synthesis/modeling/models/conditional_dalle.py:35
Methodrank
(self)
image_synthesis/data/utils/comm.py:26
Methodreconstruct
( self, input )
image_synthesis/modeling/models/unconditional_dalle.py:149
Methodreconstruct
( self, input )
image_synthesis/modeling/models/conditional_dalle.py:178
Methodreconstruct
( self, input )
image_synthesis/modeling/models/dalle.py:208
Functionreduce_dict
Args: input_dict (dict): all the values will be reduced average (bool): whether to do average or sum Reduce the values in the
image_synthesis/data/utils/comm.py:115
Functionsample
take a conditioning sequence of indices in x (of shape (b,t)) and predict the next token in the sequence, feeding the predictions back into t
image_synthesis/taming/modules/transformer/mingpt.py:292
Methodsample
( self, condition_token, condition_mask, condition_embed,
image_synthesis/modeling/transformers/diffusion_transformer.py:467
Methodsample
( self, condition_token, condition_mask, condition_embed,
image_synthesis/modeling/transformers/diffusion_transformer_amp.py:472
Methodsample_fast
( self, condition_token, condition_mask, condition_embed,
image_synthesis/modeling/transformers/diffusion_transformer_amp.py:544
Functionsample_index_randomly
x: should be 2D tensor, randomly smaple along the lat dimension
image_synthesis/modeling/utils/misc.py:51
Functionsave_dict_to_json
(d, path, indent=None)
image_synthesis/utils/io.py:18
Methodstate_dict
(self)
image_synthesis/engine/clip_grad_norm.py:26
Methodstate_dict
(self)
image_synthesis/engine/lr_scheduler.py:201
Methodstate_dict
(self)
image_synthesis/engine/lr_scheduler.py:287
Methodstem
(x)
image_synthesis/modeling/modules/clip/model.py:135
Methodstep
(self, metrics)
image_synthesis/engine/lr_scheduler.py:119
Methodstep
(self)
image_synthesis/engine/lr_scheduler.py:238
Functionsub_func
(queue,kwargs)
image_synthesis/data/utils/manage.py:18
Functionsynchronize
()
image_synthesis/distributed/distributed.py:39
Functiontorch_distributed_zero_first
Decorator to make all processes in distributed training wait for each local_master to do something.
image_synthesis/data/utils/comm.py:221
Methodtrain
(self, mode=True)
image_synthesis/modeling/embeddings/base_embedding.py:13
Methodtraining_step
(self, batch, batch_idx, optimizer_idx)
image_synthesis/taming/models/vqgan.py:83
Methodtraining_step
(self, batch, batch_idx, optimizer_idx)
image_synthesis/taming/models/vqgan.py:211
Methodtraining_step
(self, batch, batch_idx)
image_synthesis/taming/models/cond_transformer.py:289
Methodvalidate
(self)
image_synthesis/engine/solver.py:537
Methodvalidation_step
(self, batch, batch_idx)
image_synthesis/taming/models/vqgan.py:104
Methodvalidation_step
(self, batch, batch_idx)
image_synthesis/taming/models/vqgan.py:232
Methodvalidation_step
(self, batch, batch_idx)
image_synthesis/taming/models/cond_transformer.py:294
Functionvanilla_d_loss
(logits_real, logits_fake)
image_synthesis/taming/modules/losses/vqperceptual.py:27
Functionweights_init
(m)
image_synthesis/taming/modules/discriminator/model.py:8
Methodworld_size
(self)
image_synthesis/data/utils/comm.py:18
Methodwrapper
(*args, **kwards)
image_synthesis/modeling/models/unconditional_dalle.py:81
Methodwrapper
(*args, **kwards)
image_synthesis/modeling/models/conditional_dalle.py:95
Methodwrapper
(*args, **kwards)
image_synthesis/modeling/models/dalle.py:97
← previous501–543 of 543, ranked by callers