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) 


950
951
952
953
 
# File 'ext/llama_cpp/llama_cpp.c', line 950

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) 


955
956
957
958
959
 
# File 'ext/llama_cpp/llama_cpp.c', line 955

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;
}

#ftype ⇒ Integer

Returns:

  • (Integer) 


917
918
919
920
 
# File 'ext/llama_cpp/llama_cpp.c', line 917

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) 


922
923
924
925
926
 
# File 'ext/llama_cpp/llama_cpp.c', line 922

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) 


994
995
996
997
 
# File 'ext/llama_cpp/llama_cpp.c', line 994

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) 


999
1000
1001
1002
1003
 
# File 'ext/llama_cpp/llama_cpp.c', line 999

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) 


906
907
908
909
 
# File 'ext/llama_cpp/llama_cpp.c', line 906

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) 


911
912
913
914
915
 
# File 'ext/llama_cpp/llama_cpp.c', line 911

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) 


972
973
974
975
 
# File 'ext/llama_cpp/llama_cpp.c', line 972

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) 


977
978
979
980
981
 
# File 'ext/llama_cpp/llama_cpp.c', line 977

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) 


928
929
930
931
 
# File 'ext/llama_cpp/llama_cpp.c', line 928

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) 


933
934
935
936
937
 
# File 'ext/llama_cpp/llama_cpp.c', line 933

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) 


983
984
985
986
 
# File 'ext/llama_cpp/llama_cpp.c', line 983

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) 


988
989
990
991
992
 
# File 'ext/llama_cpp/llama_cpp.c', line 988

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) 


961
962
963
964
 
# File 'ext/llama_cpp/llama_cpp.c', line 961

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) 


966
967
968
969
970
 
# File 'ext/llama_cpp/llama_cpp.c', line 966

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) 


939
940
941
942
 
# File 'ext/llama_cpp/llama_cpp.c', line 939

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) 


944
945
946
947
948
 
# File 'ext/llama_cpp/llama_cpp.c', line 944

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;
}