MCPcopy Create free account

hub / github.com/zai-org/CodeGeeX / functions

Functions1,098 in github.com/zai-org/CodeGeeX

↓ 1 callersFunctionget_parameters_in_billions
(model)
codegeex/megatron/utils.py:202
↓ 1 callersFunctionget_shrink_embedding_gradient_alpha
(iteration)
codegeex/megatron/model/language_model.py:29
↓ 1 callersFunctionget_train_valid_test_split_
Get dataset splits from comma or '/' separated string list.
codegeex/megatron/data/dataset_utils.py:450
↓ 1 callersFunctioninfer_dataset_impl
(path)
codegeex/megatron/data/indexed_dataset.py:37
↓ 1 callersFunctioninitialize_wandb_experiment
Initialize wandb experiment.
codegeex/megatron/initialize.py:314
↓ 1 callersFunctionis_code_generation_finished
Checks whether the generated code text is finished.
codegeex/mindspore/src/generate_humaneval.py:25
↓ 1 callersMethodis_in_use
Whether the current buffer hold on to any memory.
codegeex/megatron/memory.py:82
↓ 1 callersFunctionis_support_nz_non_aligned
is_support_nz_non_aligned
codegeex/mindspore/scripts/layer_norm.py:709
↓ 1 callersFunctionlayer_norm_compute
DSL description of the layernorm operator's mathematical calculation process Parameters ---------- input_x: TVM tensor the p
codegeex/mindspore/scripts/layer_norm.py:571
↓ 1 callersFunctionlayer_norm_compute_nz
DSL description of the layernorm operator's mathematical calculation process Parameters ---------- input_x: TVM tensor the p
codegeex/mindspore/scripts/layer_norm.py:433
↓ 1 callersFunctionlayer_norm_x_back_nz_compute
DSL description of the layernorm_grad operator's mathematical calculation process Parameters ---------- data_dy: TVM tensor
codegeex/mindspore/scripts/layer_norm_x_backprop_v2.py:994
↓ 1 callersFunctionlayer_norm_x_backprop_v2_compute
DSL description of the layernorm_grad operator's mathematical calculation process Parameters ---------- input_dy : dict
codegeex/mindspore/scripts/layer_norm_x_backprop_v2.py:748
↓ 1 callersFunctionloadAttentionLayerFromNp
(npCkptPath, transformer, layerID)
codegeex/megatron/mindspore_to_megatron.py:90
↓ 1 callersFunctionloadEmbeddingFromNp
(npCkptPath, languageModel, vocabSize=52224)
codegeex/megatron/mindspore_to_megatron.py:73
↓ 1 callersFunctionloadModelFromNp
(sd, args)
codegeex/megatron/mindspore_to_megatron.py:53
↓ 1 callersFunctionloadQueryLayerFromNp
(npCkptPath, transformer)
codegeex/megatron/mindspore_to_megatron.py:199
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_batch.py:40
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_humaneval.py:42
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/convertion_1p.py:40
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation.py:40
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_values_1p.py:40
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_values.py:39
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_finetune.py:41
↓ 1 callersFunctionload_model
r""" The main function for load model
codegeex/mindspore/generation_1p.py:40
↓ 1 callersFunctionload_pretrain_dataset
(dataset_path: Union[str, List[str]])
codegeex/data/process_pretrain_dataset.py:32
↓ 1 callersFunctionloss_func
(loss_mask, output_tensor)
codegeex/megatron/tools/finetune_codegeex.py:157
↓ 1 callersFunctionmain
()
codegeex/torch/get_ckpt_qkv.py:9
↓ 1 callersFunctionmain
()
codegeex/megatron/mindspore_to_megatron.py:290
↓ 1 callersFunctionmain
()
codegeex/megatron/convert_ckpt_parallel.py:48
↓ 1 callersFunctionmain
()
codegeex/megatron/merge_ckpt_parallel.py:43
↓ 1 callersFunctionmain
()
codegeex/megatron/tools/collect_env.py:7
↓ 1 callersFunctionmain
()
codegeex/benchmark/gather_output.py:51
↓ 1 callersFunctionmain
()
codegeex/benchmark/inspect_result.py:281
↓ 1 callersFunctionmain
()
codegeex/benchmark/evaluate_humaneval_x.py:234
↓ 1 callersFunctionmain
()
codegeex/benchmark/humaneval-x/evaluate_humaneval_x.py:234
↓ 1 callersFunctionmain
(opt)
codegeex/paddle/pt_to_pdparams.py:47
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_batch.py:298
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_humaneval.py:245
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/convertion_1p.py:246
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation.py:249
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_values_1p.py:243
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_values.py:208
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_finetune.py:285
↓ 1 callersFunctionmain
Main process for predict or export model
codegeex/mindspore/generation_1p.py:242
↓ 1 callersFunctionmain
()
tests/test_inference_paddle.py:129
↓ 1 callersFunctionmain
()
tests/test_inference.py:118
↓ 1 callersFunctionmain
()
tests/test_inference_oneflow.py:119
↓ 1 callersFunctionmain
()
tests/test_inference_megatron.py:136
↓ 1 callersFunctionmain
()
api/codegeex-api-example-python/generation_example.py:33
↓ 1 callersFunctionmain
()
deployment/server_gradio.py:85
↓ 1 callersFunctionmake_mmap_builder
(out_file, vocab_size=None)
codegeex/megatron/data/indexed_dataset.py:119
↓ 1 callersMethodmap_token
(self, token: Union[int, str])
codegeex/mindspore/src/code_tokenizer.py:118
↓ 1 callersMethodmap_tokens
(self, tokens)
codegeex/mindspore/src/code_tokenizer.py:123
↓ 1 callersFunctionmaybe_copy
(attribute)
codegeex/megatron/mpu/layers.py:75
↓ 1 callersFunctionmaybe_set
(attribute, value)
codegeex/megatron/mpu/layers.py:66
↓ 1 callersFunctionmodel_provider
Build the model.
tests/test_inference_paddle.py:15
↓ 1 callersFunctionmodel_provider
Build the model.
tests/test_inference.py:12
↓ 1 callersFunctionmodel_provider
Build the model.
tests/test_inference_oneflow.py:17
↓ 1 callersFunctionmodel_provider
Build the model.
deployment/server_gradio.py:14
↓ 1 callersFunctionnuclear_sample_tokens
(model, input_tokens, attention_mask, position_ids, contex
codegeex/megatron/code_generation_utils.py:600
↓ 1 callersFunctionnuclear_sampling
(logits: torch.FloatTensor, temperature: float, top_p: float = None, top_k: int = None)
codegeex/megatron/code_generation_utils.py:566
↓ 1 callersFunctionnz_non_aligned
DSL description of the layernorm operator's mathematical calculation process for non_aligned scene
codegeex/mindspore/scripts/layer_norm.py:354
↓ 1 callersFunctionpackage_file
package multiple files
codegeex/mindspore/src/preprocess.py:38
↓ 1 callersFunctionpad_batch
(batch, pad_id, seq_length)
codegeex/torch/inference.py:117
↓ 1 callersFunctionpad_batch
(batch, pad_id, seq_length)
codegeex/oneflow/inference.py:117
↓ 1 callersFunctionpad_batch
(batch, pad_id, seq_length)
codegeex/paddle/inference.py:117
↓ 1 callersMethodpad_seq
(self, prompt_tokens: List[int], label: int, extra: dict = None)
codegeex/data/processor.py:114
↓ 1 callersFunctionparse_args
Parse all arguments.
codegeex/megatron/arguments.py:25
↓ 1 callersFunctionparse_opt
()
codegeex/paddle/pt_to_pdparams.py:29
↓ 1 callersMethodperformHttpPost
发起 HTTP POST 请求。 @param url 请求的URL @param payload 请求的JSON数据 @return 请求返回的内容,若出错则返回 null。
api/codegeex-api-example-java/src/main/java/cn/aminer/codegeex/example/CodeGenerationExample.java:50
↓ 1 callersFunctionprocess_extra_prompt
Processes the extra prompt.
codegeex/benchmark/utils.py:102
↓ 1 callersFunctionprocess_sample
( sample: Dict, language: str=None, mode: str="pretrain", )
codegeex/data/process_pretrain_dataset.py:49
↓ 1 callersMethodprocess_sample
Process a sample.
codegeex/data/processor.py:48
↓ 1 callersMethodprocess_sample
Process a sample.
codegeex/data/processor.py:128
↓ 1 callersMethodprocess_sample_strict
Instead of processing lazily, we turn the iterable into a list.
codegeex/data/processor.py:66
↓ 1 callersMethodprocess_sample_strict
Instead of processing lazily, we turn the iterable into a list.
codegeex/data/processor.py:143
↓ 1 callersFunctionquantize_oneflow
Replace fp16 linear with quantized linear
codegeex/quantization/quantize_oneflow.py:103
↓ 1 callersMethodread_index
(self, path)
codegeex/megatron/data/indexed_dataset.py:159
↓ 1 callersFunctionread_translation_dataset
( data_file_src: str = None, data_file_tgt: str = None, lang_src: str = None, lang_tgt: str =
codegeex/benchmark/utils.py:69
↓ 1 callersMethodreadline
(self, *args, **kwargs)
codegeex/benchmark/execution.py:447
↓ 1 callersFunctionreliability_guard
This disables various destructive functions and prevents the generated code from interfering with the test (e.g. fork bomb, killing other pro
codegeex/benchmark/execution.py:477
↓ 1 callersMethodreload_model_params
(self)
codegeex/megatron/optimizer/optimizer.py:498
↓ 1 callersFunctionreport_memory
Simple GPU memory report.
codegeex/megatron/utils.py:92
↓ 1 callersMethodreset
Reset timer.
codegeex/megatron/global_vars.py:215
↓ 1 callersFunctionrestore_checkpoint
r""" Load checkpoint process.
codegeex/mindspore/train.py:333
↓ 1 callersFunctionrestore_exception_checkpoint
Restore exception checkpoint. Args: args_param: training job params sink_size: training job sink size dataset: datase
codegeex/mindspore/train.py:425
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_batch.py:193
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_humaneval.py:195
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/convertion_1p.py:206
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation.py:195
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_values_1p.py:201
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_values.py:174
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_finetune.py:196
↓ 1 callersFunctionrun_predict
run predict
codegeex/mindspore/generation_1p.py:201
↓ 1 callersFunctionrun_train
r"""The main training process.
codegeex/mindspore/train.py:121
↓ 1 callersFunctionrun_train
r"""The main training process.
codegeex/mindspore/finetune.py:121
↓ 1 callersFunctionrun_train_pipeline
r"""The main training process in pipeline.
codegeex/mindspore/train.py:506
↓ 1 callersFunctionrun_transform_model_parallel_ckpt
(args_opt)
codegeex/mindspore/save_1p_ckpt_from_8p_ckpt.py:213
↓ 1 callersFunctionrun_transform_opt_shard_ckpt
(args_opt)
codegeex/mindspore/save_8p_ckpt.py:154
↓ 1 callersFunctionsample_sequence_batch
( model, tokenizer, context_tokens, context_lengths, attention_mask,
codegeex/torch/inference.py:217
← previousnext →401–500 of 1,098, ranked by callers