MCPcopy Create free account

hub / github.com/Facico/Chinese-Vicuna / functions

Functions461 in github.com/Facico/Chinese-Vicuna

⨍ Functions 461◇ Types & classes 80↳ Endpoints 1
↓ 69 callersMethodfrom_pretrained
(cls, model, model_id, adapter_name="default", is_trainable=False, **kwargs)
utils.py:745
↓ 59 callersFunctionprintf
(*args,**kargs)
utils.py:33
↓ 53 callersFunctionggml_dup_tensor
tools/Vicuna.cpp/ggml.c:2693
↓ 34 callersFunctionggml_nelements
tools/Vicuna.cpp/ggml.c:2251
↓ 27 callersFunctionggml_new_tensor_1d
tools/Vicuna.cpp/ggml.c:2630
↓ 26 callersFunctionggml_are_same_shape
tools/Vicuna.cpp/ggml.c:2331
↓ 23 callersFunctionggml_view_tensor
tools/Vicuna.cpp/ggml.c:3003
↓ 18 callersFunctionggml_is_contiguous
tools/Vicuna.cpp/ggml.c:2312
↓ 18 callersFunctionggml_nrows
tools/Vicuna.cpp/ggml.c:2257
↓ 18 callersFunctionggml_time_us
tools/Vicuna.cpp/ggml.c:300
↓ 16 callersFunctionggml_nbytes
tools/Vicuna.cpp/ggml.c:2263
↓ 16 callersMethodpack
(self, linear, scales, zeros, g_idx=None)
tools/quant/quant_linear.py:342
↓ 15 callersFunctionggml_add_impl
tools/Vicuna.cpp/ggml.c:3047
↓ 14 callersFunctionggml_type_sizef
tools/Vicuna.cpp/ggml.c:2277
↓ 14 callersMethodupdate
(self, target_layer)
tools/application/chatglm_lora_test.py:46
↓ 13 callersFunctionatomic_load
tools/Vicuna.cpp/ggml.c:40
↓ 13 callersFunctionggml_get_f32_1d
tools/Vicuna.cpp/ggml.c:2908
↓ 13 callersFunctionggml_vec_scale_f32
inline static void ggml_vec_scale_f32(const int n, float * y, const float v) { for (int i = 0; i < n; ++i) y[i] *= v; }
tools/Vicuna.cpp/ggml.c:1875
↓ 12 callersFunctionggml_vec_cpy_f32
tools/Vicuna.cpp/ggml.c:1218
↓ 12 callersFunctionggml_vec_dot_f32
tools/Vicuna.cpp/ggml.c:1223
↓ 11 callersFunctionggml_mul
tools/Vicuna.cpp/ggml.c:3152
↓ 11 callersFunctionggml_mul_mat
tools/Vicuna.cpp/ggml.c:3658
↓ 11 callersFunctionggml_new_tensor
tools/Vicuna.cpp/ggml.c:2622
↓ 11 callersFunctionggml_new_tensor_2d
tools/Vicuna.cpp/ggml.c:2637
↓ 10 callersFunctionatomic_store
tools/Vicuna.cpp/ggml.c:37
↓ 8 callersFunctionatomic_fetch_add
tools/Vicuna.cpp/ggml.c:43
↓ 8 callersFunctionggml_new_tensor_impl
tools/Vicuna.cpp/ggml.c:2502
↓ 7 callersFunctionatomic_fetch_sub
tools/Vicuna.cpp/ggml.c:46
↓ 7 callersMethodformat
(self, record)
utils.py:59
↓ 7 callersFunctionfp32_from_bits
tools/Vicuna.cpp/ggml.c:169
↓ 7 callersFunctionggml_compute_forward_mul_mat_use_blas
helper function to determine if it is better to use BLAS or not for large matrices, BLAS is faster
tools/Vicuna.cpp/ggml.c:5530
↓ 7 callersFunctionggml_type_size
tools/Vicuna.cpp/ggml.c:2273
↓ 7 callersFunctionggml_vec_dot_f16
tools/Vicuna.cpp/ggml.c:1260
↓ 6 callersFunctionggml_repeat
tools/Vicuna.cpp/ggml.c:3323
↓ 6 callersFunctionggml_sub_impl
tools/Vicuna.cpp/ggml.c:3086
↓ 6 callersFunctionggml_up32
tools/Vicuna.cpp/ggml.c:2352
↓ 6 callersFunctionggml_vec_mad_f32
tools/Vicuna.cpp/ggml.c:1683
↓ 5 callersFunctionfind_layers
(module, layers=[nn.Conv2d, nn.Linear], name='')
tools/quant_llama.py:11
↓ 5 callersFunctionggml_blck_size
tools/Vicuna.cpp/ggml.c:2269
↓ 5 callersFunctionggml_element_size
tools/Vicuna.cpp/ggml.c:2281
↓ 5 callersFunctionggml_graph_compute
tools/Vicuna.cpp/ggml.c:9233
↓ 5 callersFunctionggml_is_scalar
tools/Vicuna.cpp/ggml.c:2285
↓ 5 callersFunctionggml_set_f32
tools/Vicuna.cpp/ggml.c:2762
↓ 5 callersFunctionggml_vec_norm_f32
tools/Vicuna.cpp/ggml.c:1904
↓ 5 callersFunctionllama_tokenize
tools/Vicuna.cpp/utils.cpp:280
↓ 5 callersMethodpostprocess
(self, text, render=False)
prompt.py:187
↓ 4 callersMethodfind_params
(self, x, weight=False)
tools/quant/quantizer.py:34
↓ 4 callersFunctionfp32_to_bits
tools/Vicuna.cpp/ggml.c:178
↓ 4 callersFunctionggml_compute_forward
tools/Vicuna.cpp/ggml.c:8565
↓ 4 callersFunctionggml_cpy
tools/Vicuna.cpp/ggml.c:3751
↓ 4 callersFunctionggml_cycles_per_ms
tools/Vicuna.cpp/ggml.c:324
↓ 4 callersFunctionggml_free
tools/Vicuna.cpp/ggml.c:2459
↓ 4 callersFunctionggml_graph_get_parent
tools/Vicuna.cpp/ggml.c:9748
↓ 4 callersFunctionggml_graph_reset
tools/Vicuna.cpp/ggml.c:9682
↓ 4 callersFunctionggml_init
tools/Vicuna.cpp/ggml.c:2372
↓ 4 callersFunctionggml_permute
tools/Vicuna.cpp/ggml.c:3897
↓ 4 callersFunctionggml_vec_set_f32
tools/Vicuna.cpp/ggml.c:1217
↓ 4 callersFunctionggml_vec_sub_f32
tools/Vicuna.cpp/ggml.c:1216
↓ 4 callersFunctionggml_vec_sum_f32
tools/Vicuna.cpp/ggml.c:1972
↓ 4 callersFunctionggml_view_1d
tools/Vicuna.cpp/ggml.c:3847
↓ 4 callersMethodquantize
(self, x)
tools/quant/quantizer.py:117
↓ 3 callersMethodconfigure
(self, bits, perchannel=False, sym=True, mse=False, norm=2.4, grid=100, maxshrink=.8, trits=False)
tools/quant/quantizer.py:15
↓ 3 callersFunctionggml_build_forward_expand
tools/Vicuna.cpp/ggml.c:9044
↓ 3 callersFunctionggml_build_forward_impl
tools/Vicuna.cpp/ggml.c:9024
↓ 3 callersFunctionggml_can_mul_mat
tools/Vicuna.cpp/ggml.c:2303
↓ 3 callersFunctionggml_critical_section_end
TODO: make this somehow automatically executed some sort of "sentry" mechanism
tools/Vicuna.cpp/ggml.c:2227
↓ 3 callersFunctionggml_div
tools/Vicuna.cpp/ggml.c:3195
↓ 3 callersFunctionggml_fp16_to_fp32
note: do not use these inside ggml.c these are meant to be used via the ggml.h API
tools/Vicuna.cpp/ggml.c:276
↓ 3 callersFunctionggml_is_matrix
tools/Vicuna.cpp/ggml.c:2297
↓ 3 callersFunctionggml_new_f32
tools/Vicuna.cpp/ggml.c:2680
↓ 3 callersFunctionggml_opt_get_grad
tools/Vicuna.cpp/ggml.c:9899
↓ 3 callersFunctionggml_opt_set_params
tools/Vicuna.cpp/ggml.c:9877
↓ 3 callersFunctionggml_rms_norm
tools/Vicuna.cpp/ggml.c:3644
↓ 3 callersFunctionggml_up
tools/Vicuna.cpp/ggml.c:2360
↓ 3 callersFunctionggml_vec_acc_f32
tools/Vicuna.cpp/ggml.c:1214
↓ 3 callersFunctionggml_vec_add_f32
tools/Vicuna.cpp/ggml.c:1213
↓ 3 callersFunctionggml_vec_max_f32
tools/Vicuna.cpp/ggml.c:1984
↓ 3 callersFunctionggml_vec_neg_f32
tools/Vicuna.cpp/ggml.c:1219
↓ 3 callersFunctionreplace
tools/Vicuna.cpp/utils.cpp:126
↓ 3 callersMethodstream_generate
( self, input_ids: Optional[torch.Tensor] = None, generation_config: Optional[Generati
utils.py:109
↓ 2 callersMethod_quantize
(self, x, scale, zero, maxq)
tools/quant/quantizer.py:28
↓ 2 callersFunctionbytesFromNibbles
Unpack 32 4-bit fields into 32 bytes The output vector contains 32 bytes, each one in [ 0 .. 15 ] interval
tools/Vicuna.cpp/ggml.c:367
↓ 2 callersFunctiondequantize_row_q4_0
TODO: vectorize
tools/Vicuna.cpp/ggml.c:651
↓ 2 callersFunctiondequantize_row_q4_1
tools/Vicuna.cpp/ggml.c:686
↓ 2 callersFunctionget_loaders
(name, nsamples=128, seed=0, seqlen=2048, model='')
tools/datautils.py:179
↓ 2 callersFunctionggml_abs_impl
tools/Vicuna.cpp/ggml.c:3351
↓ 2 callersFunctionggml_add
tools/Vicuna.cpp/ggml.c:3070
↓ 2 callersFunctionggml_can_repeat
check if t1 can be represented as a repeatition of t0
tools/Vicuna.cpp/ggml.c:2342
↓ 2 callersFunctionggml_compute_forward_dup
tools/Vicuna.cpp/ggml.c:4433
↓ 2 callersFunctionggml_cpy_impl
tools/Vicuna.cpp/ggml.c:3726
↓ 2 callersFunctionggml_critical_section_start
barrier via spin lock
tools/Vicuna.cpp/ggml.c:2214
↓ 2 callersFunctionggml_cycles
tools/Vicuna.cpp/ggml.c:320
↓ 2 callersFunctionggml_div_impl
tools/Vicuna.cpp/ggml.c:3168
↓ 2 callersFunctionggml_dup_impl
tools/Vicuna.cpp/ggml.c:3013
↓ 2 callersFunctionggml_fp32_to_fp16
tools/Vicuna.cpp/ggml.c:280
↓ 2 callersFunctionggml_gelu_f32
tools/Vicuna.cpp/ggml.c:1915
↓ 2 callersFunctionggml_gelu_impl
tools/Vicuna.cpp/ggml.c:3522
↓ 2 callersFunctionggml_graph_dump_dot
tools/Vicuna.cpp/ggml.c:9760
↓ 2 callersFunctionggml_graph_print
tools/Vicuna.cpp/ggml.c:9692
↓ 2 callersFunctionggml_mul_impl
tools/Vicuna.cpp/ggml.c:3125
next →1–100 of 461, ranked by callers