MCPcopy Create free account

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

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

Method__init__
Batch size ramp up. Over steps = (global-batch-size - start-batch-size) / batch_size_increment increment batch size from sta
codegeex/megatron/microbatches.py:104
Method__init__
( self, optimizer, max_lr, min_lr, warmup_steps, decay_steps,
codegeex/megatron/learning_rates.py:26
Method__init__
(self, module)
codegeex/megatron/model/distributed.py:54
Method__init__
( self, module, accumulate_allreduce_grads_in_fp32, use_contiguous_buffers )
codegeex/megatron/model/distributed.py:96
Method__init__
(self, num_tokentypes=0, parallel_output=True)
codegeex/megatron/model/codegeex_model.py:130
Method__init__
(self, hidden_size, vocab_size, max_sequence_length,
codegeex/megatron/model/language_model.py:321
Method__init__
(self, init_method, output_layer_init_method, num_tokentype
codegeex/megatron/model/language_model.py:500
Method__init__
(self, init_method, output_layer_init_method, layer_number)
codegeex/megatron/model/transformer.py:110
Method__init__
(self, init_method, output_layer_init_method, layer_number)
codegeex/megatron/model/transformer.py:332
Method__init__
(self, init_method, output_layer_init_method, layer_number)
codegeex/megatron/model/transformer.py:580
Method__init__
(self, init_method, output_layer_init_method, layer_number)
codegeex/megatron/model/transformer.py:739
Method__init__
(self, init_method, output_layer_init_method)
codegeex/megatron/model/transformer.py:898
Method__init__
(self, share_word_embeddings=True)
codegeex/megatron/model/module.py:39
Method__init__
(self, datasets, weights)
codegeex/megatron/data/blendable_dataset.py:26
Method__init__
Args: name: name of the dataset. data_prefix: prefix of the data. documents: list of document indices.
codegeex/megatron/data/prompt_dataset.py:180
Method__init__
(self, path)
codegeex/megatron/data/indexed_dataset.py:234
Method__init__
(self, out_file, dtype=np.int32)
codegeex/megatron/data/indexed_dataset.py:296
Method__init__
(self, path, skip_warmup=False)
codegeex/megatron/data/indexed_dataset.py:406
Method__init__
(self, path, skip_warmup=False)
codegeex/megatron/data/indexed_dataset.py:465
Method__init__
(self, out_file, dtype=np.int64)
codegeex/megatron/data/indexed_dataset.py:565
Method__init__
( self, total_samples, consumed_samples, micro_batch_size, data_parall
codegeex/megatron/data/data_samplers.py:63
Method__init__
( self, total_samples, consumed_samples, micro_batch_size, data_parall
codegeex/megatron/data/data_samplers.py:124
Method__init__
( self, vocab_file, merges_file, errors="replace", special_tokens=None
codegeex/megatron/tokenizer/gpt2_tokenization.py:183
Method__init__
(self, vocab_file, merge_file)
codegeex/megatron/tokenizer/tokenizer.py:185
Method__init__
( self, tokenizer, ws_start: int = None, ws_len: int = None,
codegeex/megatron/tokenizer/tokenizer.py:220
Method__init__
(self, num_embeddings, embedding_dim, init_method=init.xavier_normal_)
codegeex/megatron/mpu/layers.py:152
Method__init__
( self, input_size, output_size, bias=True, gather_output=True,
codegeex/megatron/mpu/layers.py:264
Method__init__
(self)
codegeex/megatron/mpu/random.py:148
Method__init__
Input optimizer is the base optimizer for example Adam.
codegeex/megatron/optimizer/optimizer.py:63
Method__init__
( self, optimizer, clip_grad, log_num_zeros_in_grad, params_have_main_
codegeex/megatron/optimizer/optimizer.py:170
Method__init__
Initialize scale value with the input initial scale.
codegeex/megatron/optimizer/grad_scaler.py:25
Method__init__
"Grad scaler with dynamic scale that gets adjusted during training.
codegeex/megatron/optimizer/grad_scaler.py:63
Method__init__
( self, in_features: int, out_features: int, weight_bit_width: int,
codegeex/quantization/quantize_oneflow.py:54
Method__init__
( self, input_size: int, output_size: int, weight_bit_width: int, wei
codegeex/quantization/quantize.py:78
Method__init__
( self, input_size: int, output_size: int, weight_bit_width: int, wei
codegeex/quantization/quantize.py:137
Method__init__
( self, hidden_size, num_attention_heads, layer_number, fp16=True,
codegeex/oneflow/codegeex_model.py:58
Method__init__
( self, hidden_size, num_attention_heads, layer_number, fp16=True,
codegeex/oneflow/codegeex_model.py:285
Method__init__
( self, hidden_size, num_attention_heads, layer_number, layernorm_ep
codegeex/oneflow/codegeex_model.py:500
Method__init__
( self, hidden_size, num_attention_heads, layer_number, layernorm_eps
codegeex/oneflow/codegeex_model.py:579
Method__init__
( self, hidden_size, num_attention_heads, num_layers, layernorm_epsilo
codegeex/oneflow/codegeex_model.py:660
Method__init__
( self, hidden_size, vocab_size, max_sequence_length, )
codegeex/oneflow/codegeex_model.py:782
Method__init__
( self, hidden_size, vocab_size, max_sequence_length, )
codegeex/oneflow/codegeex_model.py:861
Method__init__
( self, hidden_size, num_layers, num_attention_heads, padded_vocab_siz
codegeex/oneflow/codegeex_model.py:926
Method__init__
( self, hidden_size, num_layers, num_attention_heads, padded_vocab_siz
codegeex/oneflow/codegeex_model.py:1042
Method__init__
( self, tokenize: Callable, pad_token: int, keep_order: bool = False,
codegeex/data/processor.py:9
Method__init__
( self, hidden_size, num_attention_heads, layer_number, fp16=True,
codegeex/paddle/codegeex_model.py:57
Method__init__
( self, hidden_size, num_attention_heads, layer_number, fp16=True,
codegeex/paddle/codegeex_model.py:233
Method__init__
( self, hidden_size, num_attention_heads, layer_number, layernorm_ep
codegeex/paddle/codegeex_model.py:407
Method__init__
( self, hidden_size, num_attention_heads, layer_number, layernorm_eps
codegeex/paddle/codegeex_model.py:485
Method__init__
( self, hidden_size, num_attention_heads, num_layers, layernorm_epsilo
codegeex/paddle/codegeex_model.py:566
Method__init__
( self, hidden_size, vocab_size, max_sequence_length, )
codegeex/paddle/codegeex_model.py:688
Method__init__
( self, hidden_size, vocab_size, max_sequence_length, )
codegeex/paddle/codegeex_model.py:768
Method__init__
( self, hidden_size, num_layers, num_attention_heads, padded_vocab_siz
codegeex/paddle/codegeex_model.py:834
Method__init__
( self, hidden_size, num_layers, num_attention_heads, padded_vocab_siz
codegeex/paddle/codegeex_model.py:950
Method__init__
(self, filename: str, function_names: List[str])
codegeex/kernels/__init__.py:12
Method__init__
( self, tokenizer: GPT2TokenizerFast = None, tokenizer_path: str = "EleutherAI/gpt-j
codegeex/tokenizer/tokenizer.py:34
Method__init__
(self, path, process_fn=None)
codegeex/mindspore/src/sat_dataset.py:19
Method__init__
( self, path, process_fn, length_per_sample=64 + 1024 + 4096,
codegeex/mindspore/src/sat_dataset.py:73
Method__init__
(self, path, process_fn, with_heads=True, **kwargs)
codegeex/mindspore/src/sat_dataset.py:109
Method__init__
(self, datasets, weights=None, skip_num=0, **kwargs)
codegeex/mindspore/src/sat_dataset.py:144
Method__init__
(self, ds)
codegeex/mindspore/src/sat_dataset.py:181
Method__init__
(self, ds, indices, block_size)
codegeex/mindspore/src/sat_dataset.py:204
Method__init__
(self, ds, start, length)
codegeex/mindspore/src/sat_dataset.py:224
Method__init__
(self, batch_size=32, seq_length=2048, vocab_size=40000,
codegeex/mindspore/src/pangu_alpha_config.py:26
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha.py:41
Method__init__
( self, batch_size, hidden_size, ffn_hidden_size,
codegeex/mindspore/src/pangu_alpha.py:112
Method__init__
( self, hidden_size, compute_type=mstype.float16, parallel_config=None )
codegeex/mindspore/src/pangu_alpha.py:232
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha.py:299
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha.py:474
Method__init__
(self, config, network, loss)
codegeex/mindspore/src/pangu_alpha.py:508
Method__init__
(self, backbone, generate=False, pad_token=6, seq_length=2048)
codegeex/mindspore/src/pangu_alpha.py:610
Method__init__
(self, config, network, loss)
codegeex/mindspore/src/pangu_alpha.py:654
Method__init__
(self, params, learning_rate=1e-3, beta1=0.9, beta2=0.999, eps=1e-6, weight_decay=0.0)
codegeex/mindspore/src/utils.py:49
Method__init__
(self, params, config, clip_norm=1.0)
codegeex/mindspore/src/utils.py:213
Method__init__
(self, learning_rate, end_learning_rate, warmup_steps,
codegeex/mindspore/src/utils.py:237
Method__init__
(self, network, optimizer, config, scale_update_cell=None, enable_global_norm=True)
codegeex/mindspore/src/pangu_alpha_wrapcell_finetune.py:181
Method__init__
(self, data_length)
codegeex/mindspore/src/metrics.py:68
Method__init__
(self, model, eval_dataset, ppl_metric, validation_loss, print_per_step=250, has_trained_step=0,
codegeex/mindspore/src/callbacks.py:102
Method__init__
(self, cache_dir, bucket, local_rank=0, has_trained_epoch=0, has_trained_step=0, syn_times=100)
codegeex/mindspore/src/callbacks.py:162
Method__init__
(self, network, optimizer, config, scale_update_cell=None, enable_global_norm=True)
codegeex/mindspore/src/pangu_alpha_wrapcell.py:181
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:40
Method__init__
( self, batch_size, hidden_size, ffn_hidden_size,
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:108
Method__init__
( self, hidden_size, compute_type=mstype.float16, parallel_config=None )
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:228
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:295
Method__init__
(self, config)
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:484
Method__init__
(self, config, network, loss)
codegeex/mindspore/src/pangu_alpha_fp16_predict.py:523
Method__init__
( self, dict_file: str, extra_token_ids: List[str] = None, pad
codegeex/mindspore/src/code_tokenizer.py:59
Method__init__
( self, tokenizer: GPT2TokenizerFast = None, start_extra_id: int = 10,
codegeex/mindspore/src/code_tokenizer.py:132
Method__init__
(self, params, learning_rate=1e-3, beta1=0.9, beta2=0.999, eps=1e-6, weight_decay=0.0, clip_n
codegeex/mindspore/src/adam.py:136
Method__init__
(self, vocab_file, model_file, max_len=None)
codegeex/mindspore/src/tokenization_jieba.py:30
Method__iter__
(self)
codegeex/megatron/data/data_samplers.py:107
Method__iter__
(self)
codegeex/megatron/data/data_samplers.py:160
Method__len__
(self)
codegeex/megatron/data/blendable_dataset.py:63
Method__len__
(self)
codegeex/megatron/data/prompt_dataset.py:229
Method__len__
(self)
codegeex/megatron/data/indexed_dataset.py:213
Method__len__
(self)
codegeex/megatron/data/indexed_dataset.py:462
Method__len__
(self)
codegeex/megatron/data/indexed_dataset.py:499
Method__len__
(self)
codegeex/megatron/data/data_samplers.py:99
Method__len__
(self)
codegeex/megatron/data/data_samplers.py:157
Method__len__
(self)
codegeex/megatron/tokenizer/gpt2_tokenization.py:212
← previousnext →601–700 of 1,098, ranked by callers