MCPcopy Create free account

hub / github.com/Jingyu6/speculative_prefill / functions

Functions156 in github.com/Jingyu6/speculative_prefill

⨍ Functions 156◇ Types & classes 26↳ Endpoints 2
Method_scrolls_metrics
(self)
eval/lm_eval_patch/scrolls/task.py:272
Method_scrolls_metrics
(self)
eval/lm_eval_patch/scrolls/task.py:316
Method_scrolls_metrics
(self)
eval/lm_eval_patch/scrolls/task.py:394
Methodaggregation
(self)
eval/lm_eval_patch/scrolls/task.py:220
Methodaggregation
(self)
eval/lm_eval_patch/scrolls/task.py:234
Functionchatglm_attn_forward_vllm
( self, hidden_states: torch.Tensor, position_ids: torch.Tensor, kv_cache: tor
eval/long_bench/minference_vllm_patch.py:191
Functionchatglm_layer_forward_vllm
( self, hidden_states: torch.Tensor, position_ids: torch.Tensor, kv_cache: tor
eval/long_bench/minference_vllm_patch.py:123
Functionchatglm_model_forward_vllm
( self, hidden_states: torch.Tensor, position_ids: torch.Tensor, kv_caches: Li
eval/long_bench/minference_vllm_patch.py:73
Functionclassification_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:89
Functionclean_up_fn
()
speculative_prefill/vllm_patch/__init__.py:39
Functioncode_sim_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:80
Methodcompute_metrics
(samples)
eval/lm_eval_patch/scrolls/task.py:211
Methodconstruct_requests
(self, doc, ctx, **kwargs)
eval/lm_eval_patch/scrolls/task.py:254
Methodconstruct_requests
(self, doc, ctx, **kwargs)
eval/lm_eval_patch/scrolls/task.py:286
Methodconstruct_requests
(self, doc, ctx, **kwargs)
eval/lm_eval_patch/scrolls/task.py:328
Methodconstruct_requests
(self, doc, ctx, **kwargs)
eval/lm_eval_patch/scrolls/task.py:407
Functioncount_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:47
Functioncreate_spec_worker
(*args, **kwargs)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:33
Methoddetermine_num_available_blocks
(self)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:108
Methoddevice
(self)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:212
Methoddoc_to_decontamination_query
(self, doc)
eval/lm_eval_patch/scrolls/task.py:157
Methoddoc_to_target
(self, doc)
eval/lm_eval_patch/scrolls/task.py:197
Methoddoc_to_text
(self, doc)
eval/lm_eval_patch/scrolls/task.py:295
Methoddoc_to_text
(self, doc)
eval/lm_eval_patch/scrolls/task.py:431
Methoddoc_to_text
(self, doc)
eval/lm_eval_patch/scrolls/task.py:468
Methodexecute_model_async
( self, execute_model_req: ExecuteModelRequest, )
speculative_prefill/vllm_patch/executor/gpu_executor.py:34
Functiongenerate
Generate completion for the request. The request should be a JSON object with the following fields: - prompt: the prompt to use for the gener
eval/ruler_server.py:53
Methodhas_test_docs
(self)
eval/lm_eval_patch/scrolls/task.py:129
Methodhas_training_docs
(self)
eval/lm_eval_patch/scrolls/task.py:123
Methodhas_validation_docs
(self)
eval/lm_eval_patch/scrolls/task.py:126
Functionhealth
Health check.
eval/ruler_server.py:47
Methodhigher_is_better
(self)
eval/lm_eval_patch/scrolls/task.py:203
Methodhigher_is_better
(self)
eval/lm_eval_patch/scrolls/task.py:237
Methodinit_device
(self)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:96
Methodinitialize_cache
( self, num_gpu_blocks: int, num_cpu_blocks: int )
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:125
Functionllama_layer_forward_vllm
( self, positions: torch.Tensor, hidden_states: torch.Tensor, kv_cache: torch.
eval/long_bench/minference_vllm_patch.py:95
Functionllama_model_forward_vllm
( self, input_ids: Optional[torch.Tensor], positions: torch.Tensor, kv_caches:
eval/long_bench/minference_vllm_patch.py:47
Methodload_model
(self)
speculative_prefill/vllm_patch/worker/look_ahead_spec_worker.py:48
Functionminference_patch_vllm_tp
(self, config_file, patch_config)
eval/long_bench/minference_vllm_patch.py:239
Functionprocess_results
(doc, results)
eval/lm_eval_patch/mmlu_generative/utils.py:4
Methodprocess_results
(self, doc, results)
eval/lm_eval_patch/scrolls/task.py:240
Methodprocess_results
(self, doc, results)
eval/lm_eval_patch/scrolls/task.py:279
Methodprocess_results
(self, doc, results)
eval/lm_eval_patch/scrolls/task.py:319
Methodprocess_results
(self, doc, results)
eval/lm_eval_patch/scrolls/task.py:404
Functionqa_f1_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:128
Functionqa_f1_zh_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:137
Methodrank
(self)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:208
Functionretrieval_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:56
Functionretrieval_zh_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:68
Functionrouge_zh_score
(prediction, ground_truth, **kwargs)
eval/long_bench/metrics.py:112
Methodshould_decontaminate
(self)
eval/lm_eval_patch/scrolls/task.py:154
Methodstart_worker_execution_loop
(self)
speculative_prefill/vllm_patch/worker/spec_prefill_worker.py:164
Methodtraining_docs
(self)
eval/lm_eval_patch/scrolls/task.py:132
Functionupdate_module
(m)
eval/long_bench/minference_vllm_patch.py:213
Methodvalidation_docs
(self)
eval/lm_eval_patch/scrolls/task.py:143
Functionvllm_attn_forward
( self, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, k
eval/long_bench/minference_vllm_patch.py:31
← previous101–156 of 156, ranked by callers