Class: LlamaCpp::LlamaModelQuantizeParams

Inherits:
Object
  • Object
show all
Defined in:
ext/llama_cpp/llama_cpp.c,
ext/llama_cpp/llama_cpp.c

Overview

“struct llama_model_quantize_params” wrapper class

Instance Method Summary collapse

Instance Method Details

#allow_requantize ⇒ Boolean

Returns:

  • (Boolean) 


1068
1069
1070
1071
 
# File 'ext/llama_cpp/llama_cpp.c', line 1068

static VALUE llama_model_quantize_params_get_allow_requantize(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->allow_requantize ? Qtrue : Qfalse;
}

#allow_requantize=(allow_requantize) ⇒ Boolean

Parameters:

  • allow_requantize (Boolean)

Returns:

  • (Boolean) 


1073
1074
1075
1076
1077
 
# File 'ext/llama_cpp/llama_cpp.c', line 1073

static VALUE llama_model_quantize_params_set_allow_requantize(VALUE self, VALUE allow_requantize) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->allow_requantize = RTEST(allow_requantize) ? true : false;
  return allow_requantize;
}

#dry_run ⇒ Boolean

Returns:

  • (Boolean) 


1123
1124
1125
1126
 
# File 'ext/llama_cpp/llama_cpp.c', line 1123

static VALUE llama_model_quantize_params_get_dry_run(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->dry_run ? Qtrue : Qfalse;
}

#dry_run=(dry_run) ⇒ Boolean

Parameters:

  • dry_run (Boolean)

Returns:

  • (Boolean) 


1128
1129
1130
1131
1132
 
# File 'ext/llama_cpp/llama_cpp.c', line 1128

static VALUE llama_model_quantize_params_set_dry_run(VALUE self, VALUE dry_run) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->dry_run = RTEST(dry_run) ? true : false;
  return dry_run;
}

#ftype ⇒ Integer

Returns:

  • (Integer) 


1035
1036
1037
1038
 
# File 'ext/llama_cpp/llama_cpp.c', line 1035

static VALUE llama_model_quantize_params_get_ftype(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return INT2NUM(data->ftype);
}

#ftype=(ftype) ⇒ Integer

Parameters:

  • ftype (Integer)

Returns:

  • (Integer) 


1040
1041
1042
1043
1044
 
# File 'ext/llama_cpp/llama_cpp.c', line 1040

static VALUE llama_model_quantize_params_set_ftype(VALUE self, VALUE ftype) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->ftype = (enum llama_ftype)NUM2INT(ftype);
  return ftype;
}

#keep_split ⇒ Boolean

Returns:

  • (Boolean) 


1112
1113
1114
1115
 
# File 'ext/llama_cpp/llama_cpp.c', line 1112

static VALUE llama_model_quantize_params_get_keep_split(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->keep_split ? Qtrue : Qfalse;
}

#keep_split=(keep_split) ⇒ Boolean

Parameters:

  • keep_split (Boolean)

Returns:

  • (Boolean) 


1117
1118
1119
1120
1121
 
# File 'ext/llama_cpp/llama_cpp.c', line 1117

static VALUE llama_model_quantize_params_set_keep_split(VALUE self, VALUE keep_split) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->keep_split = RTEST(keep_split) ? true : false;
  return keep_split;
}

#nthread ⇒ Integer

Returns:

  • (Integer) 


1024
1025
1026
1027
 
# File 'ext/llama_cpp/llama_cpp.c', line 1024

static VALUE llama_model_quantize_params_get_nthread(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return INT2NUM(data->nthread);
}

#nthread=(nthread) ⇒ Integer

Parameters:

  • nthread (Integer)

Returns:

  • (Integer) 


1029
1030
1031
1032
1033
 
# File 'ext/llama_cpp/llama_cpp.c', line 1029

static VALUE llama_model_quantize_params_set_nthread(VALUE self, VALUE nthread) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->nthread = NUM2INT(nthread);
  return nthread;
}

#only_copy ⇒ Boolean

Returns:

  • (Boolean) 


1090
1091
1092
1093
 
# File 'ext/llama_cpp/llama_cpp.c', line 1090

static VALUE llama_model_quantize_params_get_only_copy(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->only_copy ? Qtrue : Qfalse;
}

#only_copy=(only_copy) ⇒ Boolean

Parameters:

  • only_copy (Boolean)

Returns:

  • (Boolean) 


1095
1096
1097
1098
1099
 
# File 'ext/llama_cpp/llama_cpp.c', line 1095

static VALUE llama_model_quantize_params_set_only_copy(VALUE self, VALUE only_copy) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->only_copy = RTEST(only_copy) ? true : false;
  return only_copy;
}

#output_tensor_type ⇒ Integer

Returns:

  • (Integer) 


1046
1047
1048
1049
 
# File 'ext/llama_cpp/llama_cpp.c', line 1046

static VALUE llama_model_quantize_params_get_output_tensor_type(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return INT2NUM(data->output_tensor_type);
}

#output_tensor_type=(output_tensor_type) ⇒ Integer

Parameters:

  • output_tensor_type (Integer)

Returns:

  • (Integer) 


1051
1052
1053
1054
1055
 
# File 'ext/llama_cpp/llama_cpp.c', line 1051

static VALUE llama_model_quantize_params_set_output_tensor_type(VALUE self, VALUE output_tensor_type) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->output_tensor_type = (enum ggml_type)NUM2INT(output_tensor_type);
  return output_tensor_type;
}

#pure ⇒ Boolean

Returns:

  • (Boolean) 


1101
1102
1103
1104
 
# File 'ext/llama_cpp/llama_cpp.c', line 1101

static VALUE llama_model_quantize_params_get_pure(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->pure ? Qtrue : Qfalse;
}

#pure=(pure) ⇒ Boolean

Parameters:

  • pure (Boolean)

Returns:

  • (Boolean) 


1106
1107
1108
1109
1110
 
# File 'ext/llama_cpp/llama_cpp.c', line 1106

static VALUE llama_model_quantize_params_set_pure(VALUE self, VALUE pure) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->pure = RTEST(pure) ? true : false;
  return pure;
}

#quantize_output_tensor ⇒ Boolean

Returns:

  • (Boolean) 


1079
1080
1081
1082
 
# File 'ext/llama_cpp/llama_cpp.c', line 1079

static VALUE llama_model_quantize_params_get_quantize_output_tensor(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return data->quantize_output_tensor ? Qtrue : Qfalse;
}

#quantize_output_tensor=(quantize_output_tensor) ⇒ Boolean

Parameters:

  • quantize_output_tensor (Boolean)

Returns:

  • (Boolean) 


1084
1085
1086
1087
1088
 
# File 'ext/llama_cpp/llama_cpp.c', line 1084

static VALUE llama_model_quantize_params_set_quantize_output_tensor(VALUE self, VALUE quantize_output_tensor) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->quantize_output_tensor = RTEST(quantize_output_tensor) ? true : false;
  return quantize_output_tensor;
}

#token_embedding_type ⇒ Integer

Returns:

  • (Integer) 


1057
1058
1059
1060
 
# File 'ext/llama_cpp/llama_cpp.c', line 1057

static VALUE llama_model_quantize_params_get_token_embedding_type(VALUE self) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  return INT2NUM(data->token_embedding_type);
}

#token_embedding_type=(token_embedding_type) ⇒ Integer

Parameters:

  • token_embedding_type (Integer)

Returns:

  • (Integer) 


1062
1063
1064
1065
1066
 
# File 'ext/llama_cpp/llama_cpp.c', line 1062

static VALUE llama_model_quantize_params_set_token_embedding_type(VALUE self, VALUE token_embedding_type) {
  llama_model_quantize_params* data = get_llama_model_quantize_params(self);
  data->token_embedding_type = (enum ggml_type)NUM2INT(token_embedding_type);
  return token_embedding_type;
}