Functions — github.com/microsoft/BitNet

hub / github.com/microsoft/BitNet / functions

Functions397 in github.com/microsoft/BitNet

⨍ Functions 397◇ Types & classes 61

↓ 1 callersMethodpart

(self, n_part: int)

utils/convert-ms-to-gguf-bitnet.py:713

↓ 1 callersMethodpart

(self, n_part: int)

utils/convert.py:713

↓ 1 callersFunctionpart_lazy

(lazy_tensor: LazyTensor, n_part: int)

utils/convert.py:938

↓ 1 callersFunctionpermutate_weight_fastest

(weight)

gpu/pack_weight.py:17

↓ 1 callersMethodpermute

(self, n_head: int, n_head_kv: int)

utils/convert-ms-to-gguf-bitnet.py:709

↓ 1 callersMethodpermute

(self, n_head: int, n_head_kv: int)

utils/convert.py:709

↓ 1 callersMethodpermute_part

(self, n_part: int, n_head: int, n_head_kv: int)

utils/convert-ms-to-gguf-bitnet.py:711

↓ 1 callersMethodpermute_part

(self, n_part: int, n_head: int, n_head_kv: int)

utils/convert.py:711

↓ 1 callersFunctionpick_output_type

(model: LazyModel, output_type_str: str | None)

utils/convert-ms-to-gguf-bitnet.py:1347

↓ 1 callersFunctionpick_output_type

(model: LazyModel, output_type_str: str | None)

utils/convert.py:1275

↓ 1 callersFunctionprepare_model

()

setup_env.py:109

↓ 1 callersFunctionpreprocess_three_weights_tl2

(M, K, weight_num, BM, BY, bm, by, weight, final_weight)

utils/convert-hf-to-gguf-bitnet.py:549

↓ 1 callersFunctionpreprocess_three_weights_tl2

(M, K, weight_num, BM, BY, bm, by, weight, final_weight)

utils/generate-dummy-bitnet-model.py:620

↓ 1 callersFunctionpreprocess_two_weights_tl2

(M, K, weight_num, BM, BY, bm, by, weight, final_weight)

utils/convert-hf-to-gguf-bitnet.py:523

↓ 1 callersFunctionpreprocess_two_weights_tl2

(M, K, weight_num, BM, BY, bm, by, weight, final_weight)

utils/generate-dummy-bitnet-model.py:577

↓ 1 callersFunctionpreprocess_weights

( w: np.ndarray, bits = 2, g = 4, )

utils/convert-ms-to-gguf-bitnet.py:723

↓ 1 callersFunctionpreprocess_weights

( w: np.ndarray, bits = 2, g = 4, )

utils/convert.py:723

↓ 1 callersFunctionpreprocess_weights_tl1

( w: np.ndarray, bits = 2, g = 4, )

utils/convert-hf-to-gguf-bitnet.py:479

↓ 1 callersFunctionpreprocess_weights_tl1

( w: np.ndarray, bits = 2, g = 4, )

utils/generate-dummy-bitnet-model.py:528

↓ 1 callersFunctionpreprocess_weights_tl2

( w: np.ndarray, bits = 2, g = 4, )

utils/convert-hf-to-gguf-bitnet.py:597

↓ 1 callersFunctionpreprocess_weights_tl2

( w: np.ndarray, bits = 2, g = 4, )

utils/generate-dummy-bitnet-model.py:699

↓ 1 callersMethodprint_summary

Print summary of all tests.

utils/test_perplexity.py:490

↓ 1 callersMethodprint_summary

Print quantization summary

utils/quantize_embeddings.py:370

↓ 1 callersFunctionprocess_tl1

(weight, BM, BY, bm, by, M, K)

utils/convert-hf-to-gguf-bitnet.py:465

↓ 1 callersMethodquant_input

(self, input)

gpu/model.py:69

↓ 1 callersMethodquant_input

(self, input)

gpu/model.py:79

↓ 1 callersFunctionquant_model

(input, output)

utils/preprocess-huggingface-bitnet.py:11

↓ 1 callersFunctionquant_weight_fp16

(weight)

utils/preprocess-huggingface-bitnet.py:5

↓ 1 callersMethodquantize

Perform single quantization Args: embedding_type: Token embedding type (uppercase format, e.g., Q6_K)

utils/quantize_embeddings.py:46

↓ 1 callersMethodquantize

(self, arr: NDArray)

utils/convert-ms-to-gguf-bitnet.py:84

↓ 1 callersMethodquantize

(self, arr: NDArray)

utils/convert.py:84

↓ 1 callersMethodquantize_embedding

Quantize model with specific embedding type. Args: embedding_type: Token embedding type (uppercase, e.g., 'Q6_K'

utils/test_perplexity.py:241

↓ 1 callersMethodrun_all_quantizations

Run all quantizations Args: types_to_quantize: List of quantization types, tuples of (embedding_type, output_suf

utils/quantize_embeddings.py:275

↓ 1 callersMethodrun_all_tests

Run perplexity tests on all datasets.

utils/test_perplexity.py:326

↓ 1 callersFunctionrun_benchmark

()

utils/e2e_benchmark.py:25

↓ 1 callersMethodrun_benchmark

Run benchmark test

utils/tune_gemm_config.py:66

↓ 1 callersFunctionrun_command

Run a system command and ensure it succeeds.

run_inference_server.py:8

↓ 1 callersFunctionrun_command

Run a system command and ensure it succeeds.

run_inference.py:8

↓ 1 callersFunctionrun_command

Run a system command and ensure it succeeds.

utils/e2e_benchmark.py:8

↓ 1 callersFunctionrun_inference

()

run_inference.py:16

↓ 1 callersFunctionrun_server

()

run_inference_server.py:16

↓ 1 callersMethodrun_tuning

Run test for all configurations

utils/tune_gemm_config.py:175

↓ 1 callersMethodsave_results

Save results to CSV file.

utils/test_perplexity.py:441

↓ 1 callersMethodsave_results

Save results to CSV file

utils/tune_gemm_config.py:153

↓ 1 callersMethodsave_results_to_csv

将benchmark结果保存到CSV文件

utils/quantize_embeddings.py:329

↓ 1 callersMethodsentencepiece_tokens

(self)

utils/convert-ms-to-gguf-bitnet.py:528

↓ 1 callersMethodsentencepiece_tokens

(self)

utils/convert.py:528

↓ 1 callersMethodset_params

(self, params: str)

utils/generate-dummy-bitnet-model.py:781

↓ 1 callersMethodset_vocab

(self)

utils/convert-hf-to-gguf-bitnet.py:79

↓ 1 callersMethodset_vocab

(self)

utils/generate-dummy-bitnet-model.py:151

↓ 1 callersFunctionsetup_gguf

()

setup_env.py:152

↓ 1 callersMethodshow

(self)

gpu/stats.py:17

↓ 1 callersFunctionsquared_relu

(x: torch.Tensor)

gpu/model.py:165

↓ 1 callersMethodtest_configuration

Test single configuration

utils/tune_gemm_config.py:110

↓ 1 callersMethodto_ggml

(self)

utils/convert-ms-to-gguf-bitnet.py:715

↓ 1 callersMethodto_ggml

(self)

utils/convert.py:715

↓ 1 callersFunctiontransform_to_i2

(x : NDArray)

utils/convert-ms-to-gguf-bitnet.py:768

↓ 1 callersFunctiontransform_to_i2

(x : NDArray)

utils/convert.py:768

↓ 1 callersMethodtransform_to_tl1

(self, x: np.ndarray)

utils/generate-dummy-bitnet-model.py:811

↓ 1 callersMethodtransform_to_tl2

(self, x: np.ndarray)

utils/generate-dummy-bitnet-model.py:817

↓ 1 callersMethodtype_for_tensor

(self, name: str, tensor: LazyTensor)

utils/convert-ms-to-gguf-bitnet.py:183

↓ 1 callersMethodtype_for_tensor

(self, name: str, tensor: LazyTensor)

utils/convert.py:183

↓ 1 callersMethodvalidate_conversion_to

(self, data_type: DataType)

utils/convert-ms-to-gguf-bitnet.py:851

↓ 1 callersMethodvalidate_conversion_to

(self, data_type: DataType)

utils/convert.py:851

↓ 1 callersFunctionweight_quant

(weight)

utils/convert-ms-to-gguf-bitnet.py:953

↓ 1 callersMethodweight_quant

(self, weight)

utils/convert-hf-to-gguf-bitnet.py:970

↓ 1 callersMethodweight_quant

(self, weight)

utils/generate-dummy-bitnet-model.py:804

↓ 1 callersMethodwrite

(self)

utils/generate-dummy-bitnet-model.py:271

↓ 1 callersMethodwrite_all

( fname_out: Path, ftype: GGMLFileType, params: Params, model: LazyModel, vocab: BaseVocab, svocab: gg

utils/convert-ms-to-gguf-bitnet.py:1314

↓ 1 callersMethodwrite_all

( fname_out: Path, ftype: GGMLFileType, params: Params, model: LazyModel, vocab: BaseVocab, svocab: gg

utils/convert.py:1245

↓ 1 callersMethodwrite_tensor_data

(self, ftype: GGMLFileType, model: LazyModel, concurrency: int)

utils/convert-ms-to-gguf-bitnet.py:1247

↓ 1 callersMethodwrite_tensor_data

(self, ftype: GGMLFileType, model: LazyModel, concurrency: int)

utils/convert.py:1178

↓ 1 callersMethodwrite_tensor_info

(self)

utils/convert-ms-to-gguf-bitnet.py:1244

↓ 1 callersMethodwrite_tensor_info

(self)

utils/convert.py:1175

↓ 1 callersMethodwrite_tensors

(self)

utils/convert-hf-to-gguf-bitnet.py:159

↓ 1 callersMethodwrite_tensors

(self)

utils/generate-dummy-bitnet-model.py:231

↓ 1 callersMethodwrite_vocab

(self)

utils/convert-hf-to-gguf-bitnet.py:206

↓ 1 callersMethodwrite_vocab

(self)

utils/generate-dummy-bitnet-model.py:278

↓ 1 callersMethodwrite_vocab_only

( fname_out: Path, params: Params, vocab: Vocab, svocab: gguf.SpecialVocab, endianess: gguf.GG

utils/convert-ms-to-gguf-bitnet.py:1283

↓ 1 callersMethodwrite_vocab_only

( fname_out: Path, params: Params, vocab: Vocab, svocab: gguf.SpecialVocab, endianess: gguf.GG

utils/convert.py:1214

Method__init__

(self, model_path, llama_perplexity_bin="../build/bin/llama-perplexity", data_dir="../data",

utils/test_perplexity.py:21

Method__init__

(self, input_model, output_dir, quantize_bin="../build/bin/llama-quantize", bench_bin="../bu

utils/quantize_embeddings.py:18

Method__init__

(self, dir_model: Path, ftype: int, fname_out: Path, is_big_endian: bool, use_temp_file: bool)

utils/convert-hf-to-gguf-bitnet.py:51

Method__init__

(self, base_path: Path)

utils/convert-ms-to-gguf-bitnet.py:415

Method__init__

(self, base_path: Path)

utils/convert-ms-to-gguf-bitnet.py:423

Method__init__

(self, base_path: Path)

utils/convert-ms-to-gguf-bitnet.py:498

Method__init__

(self, base_path: Path)

utils/convert-ms-to-gguf-bitnet.py:569

Method__init__

(self, ndarray: NDArray, i2_scale: NDArray = None)

utils/convert-ms-to-gguf-bitnet.py:782

Method__init__

(self, fname_out: Path, endianess:gguf.GGUFEndian = gguf.GGUFEndian.LITTLE)

utils/convert-ms-to-gguf-bitnet.py:1147

Method__init__

(self, path: Path)

utils/convert-ms-to-gguf-bitnet.py:1573

Method__init__

(self, dir_model: Path, ftype: int, fname_out: Path, is_big_endian: bool, use_temp_file: bool)

utils/generate-dummy-bitnet-model.py:123

Method__init__

(self, config_path, model_path, threads=16)

utils/tune_gemm_config.py:19

Method__init__

(self, base_path: Path)

utils/convert.py:415

Method__init__

(self, base_path: Path)

utils/convert.py:423

Method__init__

(self, base_path: Path)

utils/convert.py:498

Method__init__

(self, base_path: Path)

utils/convert.py:569

Method__init__

(self, ndarray: NDArray, i2_scale: NDArray = None)

utils/convert.py:782

Method__init__

(self, fname_out: Path, endianess:gguf.GGUFEndian = gguf.GGUFEndian.LITTLE)

utils/convert.py:1073

Method__init__

(self, path: Path)

utils/convert.py:1428

Method__init__

Initializes the Tokenizer with a Tiktoken model. Args: model_path (str): The path to the Tiktoken model file.

gpu/tokenizer.py:45

← previous next →201–300 of 397, ranked by callers