MCPcopy Create free account

hub / github.com/Kwai-Klear/AR-GRPO / functions

Functions650 in github.com/Kwai-Klear/AR-GRPO

⨍ Functions 650◇ Types & classes 121↳ Endpoints 2
↓ 1 callersFunctioncreate_session
()
reward_utils/geneval_client.py:10
↓ 1 callersMethodcustom_load_state_dict
(self, model_weights)
autoregressive/serve/gpt_model.py:311
↓ 1 callersFunctiondecode_n_tokens
( model, cur_token: torch.Tensor, input_pos: torch.Tensor, num_new_tokens: int, cfg_scale: float, cfg
autoregressive/models/generate.py:105
↓ 1 callersFunctiondecode_one_token
(model, x: torch.Tensor, input_pos: torch.Tensor, cfg_scale: float, cfg_flag: bool, return_logits=False, **sam
autoregressive/models/generate.py:89
↓ 1 callersMethoddetermine_num_available_blocks
Profiles the peak memory usage of the model to determine how many KV blocks may be allocated without OOMs. The engine will first cond
autoregressive/serve/worker.py:121
↓ 1 callersMethoddisable_caches
(self)
autoregressive/models/gpt.py:316
↓ 1 callersFunctiondownload
(url, local_path, chunk_size=1024)
tokenizer_lg/tokenizer_image/lpips.py:24
↓ 1 callersFunctiondrop_path
Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for E
utils/drop_path.py:4
↓ 1 callersMethodencode_request
( self, request_id: str, # pylint: disable=unused-argument prompt: Optional[str],
autoregressive/serve/llm_engine.py:326
↓ 1 callersFunctionevaluate
Evaluate given image using detected objects on the global metadata specifications. Assumptions: * Metadata combines 'include'
benchmark/reward-server/reward_server/gen_eval.py:139
↓ 1 callersFunctionevaluate
Evaluate given image using detected objects on the global metadata specifications. Assumptions: * Metadata combines 'include' clauses wit
benchmark/geneval/evaluation/evaluate_images.py:161
↓ 1 callersFunctionevaluate_batch_image
(images, prompts, client)
reward_utils/unireward.py:62
↓ 1 callersFunctionevaluate_image
(image_pils, metadatas, only_strict)
benchmark/reward-server/reward_server/gen_eval.py:266
↓ 1 callersFunctionevaluate_image
(filepath, metadata)
benchmark/geneval/evaluation/evaluate_images.py:224
↓ 1 callersFunctionevaluate_image
(prompt, image, client)
reward_utils/unireward.py:38
↓ 1 callersFunctionevaluate_reward
Evaluate given image using detected objects on the global metadata specifications. Assumptions: * Metadata combines 'include'
benchmark/reward-server/reward_server/gen_eval.py:201
↓ 1 callersMethodexecute_model
( self, seq_group_metadata_list: List[SequenceGroupMetadata], kv_caches: List[torch.Te
autoregressive/serve/model_runner.py:845
↓ 1 callersMethodexecute_model
( self, seq_group_metadata_list: Optional[List[SequenceGroupMetadata]] = None, blocks_
autoregressive/serve/worker.py:212
↓ 1 callersFunctionexists
(val)
tokenizer_lg/tokenizer_image/discriminator.py:254
↓ 1 callersFunctionexists
(val)
tokenizer_lg/tokenizer_image/discriminator_stylegan.py:100
↓ 1 callersFunctionf
(_)
benchmark/reward-server/test/test_deqa.py:14
↓ 1 callersFunctionf
(_)
benchmark/reward-server/test/test_geneval.py:15
↓ 1 callersFunctionfind_multiple
(n: int, k: int)
autoregressive/serve/gpt_model.py:18
↓ 1 callersMethodfind_prefix
(self,s,l)
autoregressive/modeling.py:210
↓ 1 callersMethodforward
( self, input_ids: torch.Tensor, positions: torch.Tensor, kv_caches: List[torc
autoregressive/serve/model_runner.py:1162
↓ 1 callersMethodfrom_engine_args
Creates an LLM engine from the engine arguments.
autoregressive/serve/llm_engine.py:255
↓ 1 callersMethodgen_fn
()
fid_evaluation.py:447
↓ 1 callersFunctiongenerate_color_attribution_sample
(rng: np.random.Generator)
benchmark/geneval/prompts/create_prompts.py:121
↓ 1 callersFunctiongenerate_color_sample
(rng: np.random.Generator)
benchmark/geneval/prompts/create_prompts.py:89
↓ 1 callersFunctiongenerate_counting_sample
(rng: np.random.Generator, max_count=4)
benchmark/geneval/prompts/create_prompts.py:70
↓ 1 callersFunctiongenerate_position_sample
(rng: np.random.Generator)
benchmark/geneval/prompts/create_prompts.py:106
↓ 1 callersFunctiongenerate_single_object_sample
(rng: np.random.Generator, size: int = None)
benchmark/geneval/prompts/create_prompts.py:33
↓ 1 callersFunctiongenerate_suite
(rng: np.random.Generator, n: int = 100, output_path: str = "")
benchmark/geneval/prompts/create_prompts.py:138
↓ 1 callersFunctiongenerate_two_object_sample
(rng: np.random.Generator)
benchmark/geneval/prompts/create_prompts.py:54
↓ 1 callersFunctionget_activations
Calculates the activations of the pool_3 layer for all images. Params: -- files : List of image files paths -- model : Instan
reward_utils/fid.py:65
↓ 1 callersMethodget_cache_block_size_bytes
Get the size of the KV cache block size in bytes.
autoregressive/serve/worker.py:274
↓ 1 callersMethodget_codebook_entry
(self, indices, shape)
tokenizer_lg/vqgan/quantize.py:92
↓ 1 callersMethodget_codebook_entry
(self, indices, shape=None, channel_first=True)
tokenizer_lg/tokenizer_image/vq_model.py:261
↓ 1 callersMethodget_max_block_per_batch
(self)
autoregressive/serve/model_runner.py:244
↓ 1 callersMethodget_model_config
Gets the model configuration.
autoregressive/serve/llm_engine.py:458
↓ 1 callersMethodget_model_device
(self)
autoregressive/modeling.py:201
↓ 1 callersMethodget_num_unfinished_requests
Gets the number of unfinished requests.
autoregressive/serve/llm_engine.py:462
↓ 1 callersMethodget_patch
(self, idx)
reward_utils/maniqa.py:42
↓ 1 callersMethodgraph
(self)
autoregressive/serve/model_runner.py:1108
↓ 1 callersMethodhas_unfinished_requests
Returns True if there are unfinished requests.
autoregressive/serve/llm_engine.py:466
↓ 1 callersFunctioninit_worker_distributed_environment
Initialize the distributed environment.
autoregressive/serve/worker.py:282
↓ 1 callersMethodinitialize_cache
Allocate GPU and CPU KV cache with the specified number of blocks. This also warms up the model, which may record CUDA graphs.
autoregressive/serve/worker.py:166
↓ 1 callersMethodinitialize_weights
(self)
autoregressive/models/gpt.py:300
↓ 1 callersMethodless_thans
(self, batch_1, radii_1, batch_2, radii_2)
fid_evaluation.py:406
↓ 1 callersFunctionload_deqascore
()
benchmark/reward-server/reward_server/deqa.py:4
↓ 1 callersMethodload_from_pretrained
(self, name="vgg_lpips")
tokenizer_lg/tokenizer_image/lpips.py:69
↓ 1 callersFunctionload_geneval
()
benchmark/reward-server/reward_server/gen_eval.py:27
↓ 1 callersFunctionload_imagenet_stat
(spath="../VAR/VIRTUAL_imagenet256_labeled.npz")
reward_utils/fid.py:253
↓ 1 callersMethodload_model
(self, model_ckpt, map_location='cpu')
autoregressive/modeling.py:180
↓ 1 callersMethodload_model
(self, args)
autoregressive/serve/worker.py:117
↓ 1 callersFunctionload_models
()
benchmark/reward-server/reward_server/gen_eval.py:40
↓ 1 callersFunctionload_models
(args)
benchmark/geneval/evaluation/evaluate_images.py:61
↓ 1 callersMethodload_vq
(self, vq_ckpt, map_location='cpu')
autoregressive/modeling.py:175
↓ 1 callersMethodlog_images
(self,outputs,labels=None)
img_gen_grpo_trainer.py:16
↓ 1 callersFunctionmain
(script_args, training_args, model_args)
img_gen_grpo_train.py:123
↓ 1 callersFunctionmain
()
fid_evaluation.py:27
↓ 1 callersFunctionmain
(args)
tokenizer_lg/vae/sd_vae_demo.py:9
↓ 1 callersFunctionmain
(args)
tokenizer_lg/vae/reconstruction_vae_ddp.py:81
↓ 1 callersFunctionmain
(args)
tokenizer_lg/vqgan/reconstruction_vqgan_ddp.py:83
↓ 1 callersFunctionmain
(args)
tokenizer_lg/vqgan/taming_vqgan_demo.py:17
↓ 1 callersFunctionmain
(args)
tokenizer_lg/validation/val_ddp.py:76
↓ 1 callersFunctionmain
(args)
tokenizer_lg/tokenizer_image/reconstruction_vq_ddp.py:43
↓ 1 callersFunctionmain
(args)
tokenizer_lg/tokenizer_image/vq_demo.py:13
↓ 1 callersFunctionmain
Trains a new model.
tokenizer_lg/tokenizer_image/vq_train.py:36
↓ 1 callersFunctionmain
(args)
tokenizer_lg/consistencydecoder/reconstruction_cd_ddp.py:81
↓ 1 callersFunctionmain
(args)
tokenizer_lg/consistencydecoder/cd_demo.py:9
↓ 1 callersFunctionmain
Trains a new DiT model.
language/extract_t5_feature.py:53
↓ 1 callersFunctionmain
(opt)
benchmark/geneval/generation/diffusers_generate.py:100
↓ 1 callersFunctionmain
(args)
benchmark/geneval/evaluation/evaluate_images.py:262
↓ 1 callersFunctionmain
()
reward_utils/fid.py:328
↓ 1 callersFunctionmain
(args)
autoregressive/serve/sample_c2i.py:12
↓ 1 callersFunctionmain
(args)
autoregressive/sample/sample_t2i.py:22
↓ 1 callersFunctionmain
(args)
autoregressive/sample/sample_c2i_ddp.py:39
↓ 1 callersFunctionmain
(args)
autoregressive/sample/sample_c2i.py:20
↓ 1 callersFunctionmain
(args)
autoregressive/sample/sample_t2i_ddp.py:26
↓ 1 callersFunctionmain
(args)
autoregressive/train/extract_codes_c2i.py:23
↓ 1 callersFunctionmain
(args)
autoregressive/train/train_c2i_fsdp.py:102
↓ 1 callersFunctionmain
Trains a new DiT model.
autoregressive/train/extract_codes_t2i.py:56
↓ 1 callersFunctionmain
(args)
autoregressive/train/train_c2i.py:57
↓ 1 callersFunctionmain
(args)
autoregressive/train/train_t2i.py:27
↓ 1 callersFunctionmake_plural
(name: str)
benchmark/geneval/prompts/create_prompts.py:26
↓ 1 callersFunctionnanmax
Compute the maximum value of a tensor, ignoring NaNs. This function only supports 1D tensors. Args: tensor (`torch.Tensor`): Input t
modified_grpo_trainer.py:271
↓ 1 callersFunctionnanmin
Compute the minimum value of a tensor, ignoring NaNs. This function only supports 1D tensors. Args: tensor (`torch.Tensor`): Input t
modified_grpo_trainer.py:256
↓ 1 callersFunctionnanstd
Compute the standard deviation of a tensor, ignoring NaNs. This function only supports 1D tensors. Args: tensor (`torch.Tensor`):
modified_grpo_trainer.py:184
↓ 1 callersFunctionopen_npz_array
(path: str, arr_name: str)
fid_evaluation.py:521
↓ 1 callersFunctionparse_args
()
benchmark/geneval/generation/diffusers_generate.py:21
↓ 1 callersFunctionparse_args
()
benchmark/geneval/evaluation/evaluate_images.py:29
↓ 1 callersFunctionpil_image_to_base64
(image)
reward_utils/unireward.py:13
↓ 1 callersFunctionpil_loader
(path)
lazy_dataset.py:55
↓ 1 callersFunctionprecompute_freqs_cis_2d
(grid_size: int, n_elem: int, base: int = 10000, cls_token_num=120)
autoregressive/serve/gpt_model.py:344
↓ 1 callersFunctionprefill
(model, cond_idx: torch.Tensor, input_pos: torch.Tensor, cfg_scale: float, return_logits=False, **sampling_kwa
autoregressive/models/generate.py:77
↓ 1 callersMethodprepare_input_tensors
( self, seq_group_metadata_list: List[SequenceGroupMetadata], )
autoregressive/serve/model_runner.py:676
↓ 1 callersMethodprocess_image
(self, img_tensor)
img_gen_grpo_rewards.py:101
↓ 1 callersMethodprocess_image
(self, img_tensor)
img_gen_grpo_rewards.py:162
↓ 1 callersMethodprocess_image
(self, img_tensor)
img_gen_grpo_rewards.py:494
← previousnext →201–300 of 650, ranked by callers