ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-01-26 17:20:01 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
733660accd813c8481c1ab5f354ca7fd701b580a
ik_llama.cpp/examples/quantize/quantize.cpp
Kawrakow 733660accd Adding ability to have meta data per tensor row (#61) 
* POC: per row scale

This is a POC how to work around opinionated ggml to
have scales per row rather than per block.
Only implemened for Zen4 and only for iq2_tn.

* POC per row scale: iq2_tn on NEON

* POC per row scale: iq2_tn on Metal

* Per row scale Metal templates

* iq1_tn: shrink to 1.625 bpw (NEON and Metal)

* POC per row scale: CUDA

* POC per row scale: add CUDA TODOs

There are two places in ggml-cuda.cu left where it is assumed
that type_size * n_per_row / block_size is the way to compute
and handle row sizes. This does not affect simple usage,
but will lead to issues when tensors are split between GPUs.

* Per row scales - CUDA

The only place left where there are unnecessary assumptions being made
is in the Flash Attention code. As we are not using any quants that
use per row scales for quantized KV cache, it should be OK for now.

* Update IQ1_TN and IQ2_TN bpw shown to user

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
2024-09-27 08:16:06 +03:00

466 lines
20 KiB
C++
Raw Blame History

#include "common.h"
#include "llama.h"
#include <cstdio>
#include <cstring>
#include <vector>
#include <string>
#include <unordered_map>
#include <fstream>
#include <cmath>
struct quant_option {
std::string name;
llama_ftype ftype;
std::string desc;
};
static const std::vector<struct quant_option> QUANT_OPTIONS = {
{ "Q4_0", LLAMA_FTYPE_MOSTLY_Q4_0, " 3.56G, +0.2166 ppl @ LLaMA-v1-7B", },
{ "Q4_1", LLAMA_FTYPE_MOSTLY_Q4_1, " 3.90G, +0.1585 ppl @ LLaMA-v1-7B", },
{ "Q5_0", LLAMA_FTYPE_MOSTLY_Q5_0, " 4.33G, +0.0683 ppl @ LLaMA-v1-7B", },
{ "Q5_1", LLAMA_FTYPE_MOSTLY_Q5_1, " 4.70G, +0.0349 ppl @ LLaMA-v1-7B", },
{ "IQ2_XXS", LLAMA_FTYPE_MOSTLY_IQ2_XXS, " 2.06 bpw quantization", },
{ "IQ2_XS", LLAMA_FTYPE_MOSTLY_IQ2_XS, " 2.31 bpw quantization", },
{ "IQ2_S", LLAMA_FTYPE_MOSTLY_IQ2_S, " 2.5 bpw quantization", },
{ "IQ2_M", LLAMA_FTYPE_MOSTLY_IQ2_M, " 2.7 bpw quantization", },
{ "IQ1_S", LLAMA_FTYPE_MOSTLY_IQ1_S, " 1.56 bpw quantization", },
{ "IQ1_M", LLAMA_FTYPE_MOSTLY_IQ1_M, " 1.75 bpw quantization", },
{ "IQ1_BN", LLAMA_FTYPE_MOSTLY_IQ1_BN, " 1.62 bpw quantization (Bitnet)", },
{ "IQ2_BN", LLAMA_FTYPE_MOSTLY_IQ2_BN, " 2.00 bpw quantization (Bitnet)", },
{ "IQ1_TN", LLAMA_FTYPE_MOSTLY_IQ1_TN, " 1.63 bpw quantization (TriLM)", },
{ "IQ2_TN", LLAMA_FTYPE_MOSTLY_IQ2_TN, " 2.00 bpw quantization (TriLM)", },
{ "Q2_K", LLAMA_FTYPE_MOSTLY_Q2_K, " 2.63G, +0.6717 ppl @ LLaMA-v1-7B", },
{ "Q2_K_S", LLAMA_FTYPE_MOSTLY_Q2_K_S, " 2.16G, +9.0634 ppl @ LLaMA-v1-7B", },
{ "IQ3_XXS", LLAMA_FTYPE_MOSTLY_IQ3_XXS, " 3.06 bpw quantization", },
{ "IQ3_S", LLAMA_FTYPE_MOSTLY_IQ3_S, " 3.44 bpw quantization", },
{ "IQ3_M", LLAMA_FTYPE_MOSTLY_IQ3_M, " 3.66 bpw quantization mix", },
{ "Q3_K", LLAMA_FTYPE_MOSTLY_Q3_K_M, "alias for Q3_K_M" },
{ "IQ3_XS", LLAMA_FTYPE_MOSTLY_IQ3_XS, " 3.3 bpw quantization" , },
{ "Q3_K_S", LLAMA_FTYPE_MOSTLY_Q3_K_S, " 2.75G, +0.5551 ppl @ LLaMA-v1-7B", },
{ "Q3_K_M", LLAMA_FTYPE_MOSTLY_Q3_K_M, " 3.07G, +0.2496 ppl @ LLaMA-v1-7B", },
{ "Q3_K_L", LLAMA_FTYPE_MOSTLY_Q3_K_L, " 3.35G, +0.1764 ppl @ LLaMA-v1-7B", },
{ "IQ4_NL", LLAMA_FTYPE_MOSTLY_IQ4_NL, " 4.50 bpw non-linear quantization", },
{ "IQ4_XS", LLAMA_FTYPE_MOSTLY_IQ4_XS, " 4.25 bpw non-linear quantization", },
{ "IQ2_K", LLAMA_FTYPE_MOSTLY_IQ2_K, " 2.375 bpw non-linear quantization",},
{ "IQ3_K", LLAMA_FTYPE_MOSTLY_IQ3_K, " 3.44 bpw non-linear quantization", },
{ "IQ4_K", LLAMA_FTYPE_MOSTLY_IQ4_K, " 4.5 bpw non-linear quantization", },
{ "IQ5_K", LLAMA_FTYPE_MOSTLY_IQ5_K, " 5.5 bpw non-linear quantization", },
{ "IQ6_K", LLAMA_FTYPE_MOSTLY_IQ6_K, " 6.6 bpw non-linear quantization", },
{ "Q4_K", LLAMA_FTYPE_MOSTLY_Q4_K_M, "alias for Q4_K_M", },
{ "Q4_K_S", LLAMA_FTYPE_MOSTLY_Q4_K_S, " 3.59G, +0.0992 ppl @ LLaMA-v1-7B", },
{ "Q4_K_M", LLAMA_FTYPE_MOSTLY_Q4_K_M, " 3.80G, +0.0532 ppl @ LLaMA-v1-7B", },
{ "Q5_K", LLAMA_FTYPE_MOSTLY_Q5_K_M, "alias for Q5_K_M", },
{ "Q5_K_S", LLAMA_FTYPE_MOSTLY_Q5_K_S, " 4.33G, +0.0400 ppl @ LLaMA-v1-7B", },
{ "Q5_K_M", LLAMA_FTYPE_MOSTLY_Q5_K_M, " 4.45G, +0.0122 ppl @ LLaMA-v1-7B", },
{ "Q6_K", LLAMA_FTYPE_MOSTLY_Q6_K, " 5.15G, +0.0008 ppl @ LLaMA-v1-7B", },
{ "Q8_0", LLAMA_FTYPE_MOSTLY_Q8_0, " 6.70G, +0.0004 ppl @ LLaMA-v1-7B", },
{ "Q4_0_4_4", LLAMA_FTYPE_MOSTLY_Q4_0_4_4, " 4.34G, +0.4685 ppl @ Llama-3-8B", },
{ "Q4_0_4_8", LLAMA_FTYPE_MOSTLY_Q4_0_4_8, " 4.34G, +0.4685 ppl @ Llama-3-8B", },
{ "Q4_0_8_8", LLAMA_FTYPE_MOSTLY_Q4_0_8_8, " 4.34G, +0.4685 ppl @ Llama-3-8B", },
{ "F16", LLAMA_FTYPE_MOSTLY_F16, "14.00G, -0.0020 ppl @ Mistral-7B", },
{ "BF16", LLAMA_FTYPE_MOSTLY_BF16, "14.00G, -0.0050 ppl @ Mistral-7B", },
{ "F32", LLAMA_FTYPE_ALL_F32, "26.00G @ 7B", },
// Note: Ensure COPY comes after F32 to avoid ftype 0 from matching.
{ "COPY", LLAMA_FTYPE_ALL_F32, "only copy tensors, no quantizing", },
};
static const char * const LLM_KV_QUANTIZE_IMATRIX_FILE = "quantize.imatrix.file";
static const char * const LLM_KV_QUANTIZE_IMATRIX_DATASET = "quantize.imatrix.dataset";
static const char * const LLM_KV_QUANTIZE_IMATRIX_N_ENTRIES = "quantize.imatrix.entries_count";
static const char * const LLM_KV_QUANTIZE_IMATRIX_N_CHUNKS = "quantize.imatrix.chunks_count";
static bool try_parse_ftype(const std::string & ftype_str_in, llama_ftype & ftype, std::string & ftype_str_out) {
std::string ftype_str;
for (auto ch : ftype_str_in) {
ftype_str.push_back(std::toupper(ch));
}
for (auto & it : QUANT_OPTIONS) {
if (it.name == ftype_str) {
ftype = it.ftype;
ftype_str_out = it.name;
return true;
}
}
try {
int ftype_int = std::stoi(ftype_str);
for (auto & it : QUANT_OPTIONS) {
if (it.ftype == ftype_int) {
ftype = it.ftype;
ftype_str_out = it.name;
return true;
}
}
}
catch (...) {
// stoi failed
}
return false;
}
// usage:
// ./llama-quantize [--allow-requantize] [--leave-output-tensor] [--pure] models/llama/ggml-model.gguf [models/llama/ggml-model-quant.gguf] type [nthreads]
//
[[noreturn]]
static void usage(const char * executable) {
printf("usage: %s [--help] [--allow-requantize] [--leave-output-tensor] [--pure] [--imatrix] [--include-weights] [--exclude-weights] [--output-tensor-type] [--token-embedding-type] [--override-kv] model-f32.gguf [model-quant.gguf] type [nthreads]\n\n", executable);
printf(" --allow-requantize: Allows requantizing tensors that have already been quantized. Warning: This can severely reduce quality compared to quantizing from 16bit or 32bit\n");
printf(" --leave-output-tensor: Will leave output.weight un(re)quantized. Increases model size but may also increase quality, especially when requantizing\n");
printf(" --pure: Disable k-quant mixtures and quantize all tensors to the same type\n");
printf(" --imatrix file_name: use data in file_name as importance matrix for quant optimizations\n");
printf(" --include-weights tensor_name: use importance matrix for this/these tensor(s)\n");
printf(" --exclude-weights tensor_name: use importance matrix for this/these tensor(s)\n");
printf(" --output-tensor-type ggml_type: use this ggml_type for the output.weight tensor\n");
printf(" --token-embedding-type ggml_type: use this ggml_type for the token embeddings tensor\n");
printf(" --keep-split: will generate quatized model in the same shards as input");
printf(" --override-kv KEY=TYPE:VALUE\n");
printf(" Advanced option to override model metadata by key in the quantized model. May be specified multiple times.\n");
printf("Note: --include-weights and --exclude-weights cannot be used together\n");
printf("\nAllowed quantization types:\n");
for (auto & it : QUANT_OPTIONS) {
if (it.name != "COPY") {
printf(" %2d or ", it.ftype);
} else {
printf(" ");
}
printf("%-7s : %s\n", it.name.c_str(), it.desc.c_str());
}
exit(1);
}
static int load_imatrix(const std::string & imatrix_file, std::string & imatrix_dataset, std::unordered_map<std::string, std::vector<float>> & imatrix_data) {
std::ifstream in(imatrix_file.c_str(), std::ios::binary);
if (!in) {
printf("%s: failed to open %s\n",__func__, imatrix_file.c_str());
exit(1);
}
int n_entries;
in.read((char *)&n_entries, sizeof(n_entries));
if (in.fail() || n_entries < 1) {
printf("%s: no data in file %s\n", __func__, imatrix_file.c_str());
exit(1);
}
for (int i = 0; i < n_entries; ++i) {
int len; in.read((char *)&len, sizeof(len));
std::vector<char> name_as_vec(len+1);
in.read((char *)name_as_vec.data(), len);
if (in.fail()) {
printf("%s: failed reading name for entry %d from %s\n", __func__, i+1, imatrix_file.c_str());
exit(1);
}
name_as_vec[len] = 0;
std::string name{name_as_vec.data()};
auto & e = imatrix_data[name];
int ncall;
in.read((char *)&ncall, sizeof(ncall));
int nval;
in.read((char *)&nval, sizeof(nval));
if (in.fail() || nval < 1) {
printf("%s: failed reading number of values for entry %d\n", __func__, i);
imatrix_data = {};
exit(1);
}
e.resize(nval);
in.read((char *)e.data(), nval*sizeof(float));
if (in.fail()) {
printf("%s: failed reading data for entry %d\n", __func__, i);
imatrix_data = {};
exit(1);
}
if (ncall > 0) {
for (auto& v : e) v /= ncall;
}
if (getenv("LLAMA_TRACE")) {
printf("%s: loaded data (size = %6d, ncall = %6d) for '%s'\n", __func__, int(e.size()), ncall, name.c_str());
}
}
// latest imatrix version contains the dataset filename at the end of the file
int m_last_call = 0;
if (in.peek() != EOF) {
in.read((char *)&m_last_call, sizeof(m_last_call));
int dataset_len;
in.read((char *)&dataset_len, sizeof(dataset_len));
std::vector<char> dataset_as_vec(dataset_len);
in.read(dataset_as_vec.data(), dataset_len);
imatrix_dataset.assign(dataset_as_vec.begin(), dataset_as_vec.end());
printf("%s: imatrix dataset='%s'\n", __func__, imatrix_dataset.c_str());
}
printf("%s: loaded %d importance matrix entries from %s computed on %d chunks\n", __func__, int(imatrix_data.size()), imatrix_file.c_str(), m_last_call);
return m_last_call;
}
static int prepare_imatrix(const std::string & imatrix_file,
std::string & imatrix_dataset,
const std::vector<std::string> & included_weights,
const std::vector<std::string> & excluded_weights,
std::unordered_map<std::string, std::vector<float>> & imatrix_data) {
int m_last_call = -1;
if (!imatrix_file.empty()) {
m_last_call = load_imatrix(imatrix_file, imatrix_dataset, imatrix_data);
}
if (imatrix_data.empty()) {
return m_last_call;
}
if (!excluded_weights.empty()) {
for (auto& name : excluded_weights) {
for (auto it = imatrix_data.begin(); it != imatrix_data.end(); ) {
auto pos = it->first.find(name);
if (pos != std::string::npos) it = imatrix_data.erase(it);
else ++it;
}
}
}
if (!included_weights.empty()) {
std::unordered_map<std::string, std::vector<float>> tmp;
for (auto& name : included_weights) {
for (auto& e : imatrix_data) {
auto pos = e.first.find(name);
if (pos != std::string::npos) {
tmp.emplace(std::move(e));
}
}
}
imatrix_data = std::move(tmp);
}
if (!imatrix_data.empty()) {
printf("%s: have %d importance matrix entries\n", __func__, int(imatrix_data.size()));
}
return m_last_call;
}
static ggml_type parse_ggml_type(const char * arg) {
ggml_type result = GGML_TYPE_COUNT;
for (int j = 0; j < GGML_TYPE_COUNT; ++j) {
auto type = ggml_type(j);
const auto * name = ggml_type_name(type);
if (name && strcmp(arg, name) == 0) {
result = type; break;
}
}
return result;
}
int main(int argc, char ** argv) {
if (argc < 3) {
usage(argv[0]);
}
llama_model_quantize_params params = llama_model_quantize_default_params();
int arg_idx = 1;
std::string imatrix_file;
std::vector<std::string> included_weights, excluded_weights;
std::vector<llama_model_kv_override> kv_overrides;
for (; arg_idx < argc && strncmp(argv[arg_idx], "--", 2) == 0; arg_idx++) {
if (strcmp(argv[arg_idx], "--leave-output-tensor") == 0) {
params.quantize_output_tensor = false;
} else if (strcmp(argv[arg_idx], "--ignore-imatrix-rules") == 0) {
params.ignore_imatrix_rules = true;
} else if (strcmp(argv[arg_idx], "--output-tensor-type") == 0) {
if (arg_idx < argc-1) {
params.output_tensor_type = parse_ggml_type(argv[++arg_idx]);
} else {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--token-embedding-type") == 0) {
if (arg_idx < argc-1) {
params.token_embedding_type = parse_ggml_type(argv[++arg_idx]);
} else {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--override-kv") == 0) {
if (arg_idx == argc-1 || !string_parse_kv_override(argv[++arg_idx], kv_overrides)) {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--allow-requantize") == 0) {
params.allow_requantize = true;
} else if (strcmp(argv[arg_idx], "--pure") == 0) {
params.pure = true;
} else if (strcmp(argv[arg_idx], "--imatrix") == 0) {
if (arg_idx < argc-1) {
imatrix_file = argv[++arg_idx];
} else {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--include-weights") == 0) {
if (arg_idx < argc-1) {
included_weights.emplace_back(argv[++arg_idx]);
} else {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--exclude-weights") == 0) {
if (arg_idx < argc-1) {
excluded_weights.emplace_back(argv[++arg_idx]);
} else {
usage(argv[0]);
}
} else if (strcmp(argv[arg_idx], "--keep-split") == 0) {
params.keep_split = true;
} else {
usage(argv[0]);
}
}
if (argc - arg_idx < 2) {
printf("%s: bad arguments\n", argv[0]);
usage(argv[0]);
}
if (!included_weights.empty() && !excluded_weights.empty()) {
usage(argv[0]);
}
std::string imatrix_dataset;
std::unordered_map<std::string, std::vector<float>> imatrix_data;
int m_last_call = prepare_imatrix(imatrix_file, imatrix_dataset, included_weights, excluded_weights, imatrix_data);
if (!imatrix_data.empty()) {
params.imatrix = &imatrix_data;
{
llama_model_kv_override kvo;
std::strcpy(kvo.key, LLM_KV_QUANTIZE_IMATRIX_FILE);
kvo.tag = LLAMA_KV_OVERRIDE_TYPE_STR;
strncpy(kvo.val_str, imatrix_file.c_str(), 127);
kvo.val_str[127] = '\0';
kv_overrides.emplace_back(std::move(kvo));
}
if (!imatrix_dataset.empty()) {
llama_model_kv_override kvo;
std::strcpy(kvo.key, LLM_KV_QUANTIZE_IMATRIX_DATASET);
kvo.tag = LLAMA_KV_OVERRIDE_TYPE_STR;
strncpy(kvo.val_str, imatrix_dataset.c_str(), 127);
kvo.val_str[127] = '\0';
kv_overrides.emplace_back(std::move(kvo));
}
{
llama_model_kv_override kvo;
std::strcpy(kvo.key, LLM_KV_QUANTIZE_IMATRIX_N_ENTRIES);
kvo.tag = LLAMA_KV_OVERRIDE_TYPE_INT;
kvo.val_i64 = imatrix_data.size();
kv_overrides.emplace_back(std::move(kvo));
}
if (m_last_call > 0) {
llama_model_kv_override kvo;
std::strcpy(kvo.key, LLM_KV_QUANTIZE_IMATRIX_N_CHUNKS);
kvo.tag = LLAMA_KV_OVERRIDE_TYPE_INT;
kvo.val_i64 = m_last_call;
kv_overrides.emplace_back(std::move(kvo));
}
}
if (!kv_overrides.empty()) {
kv_overrides.emplace_back();
kv_overrides.back().key[0] = 0;
params.kv_overrides = &kv_overrides;
}
llama_backend_init();
// parse command line arguments
const std::string fname_inp = argv[arg_idx];
arg_idx++;
std::string fname_out;
std::string ftype_str;
std::string suffix = ".gguf";
if (try_parse_ftype(argv[arg_idx], params.ftype, ftype_str)) {
std::string fpath;
const size_t pos = fname_inp.find_last_of("/\\");
if (pos != std::string::npos) {
fpath = fname_inp.substr(0, pos + 1);
}
// export as [inp path]/ggml-model-[ftype]. Only add extension if there is no splitting
fname_out = fpath + "ggml-model-" + ftype_str;
if (!params.keep_split) {
fname_out += suffix;
}
arg_idx++;
if (ftype_str == "COPY") {
params.only_copy = true;
}
} else {
fname_out = argv[arg_idx];
if (params.keep_split && fname_out.find(suffix) != std::string::npos) {
fname_out = fname_out.substr(0, fname_out.length() - suffix.length());
}
arg_idx++;
if (argc <= arg_idx) {
fprintf(stderr, "%s: missing ftype\n", __func__);
return 1;
}
if (!try_parse_ftype(argv[arg_idx], params.ftype, ftype_str)) {
fprintf(stderr, "%s: invalid ftype '%s'\n", __func__, argv[3]);
return 1;
}
if (ftype_str == "COPY") {
params.only_copy = true;
}
arg_idx++;
}
// parse nthreads
if (argc > arg_idx) {
try {
params.nthread = std::stoi(argv[arg_idx]);
}
catch (const std::exception & e) {
fprintf(stderr, "%s: invalid nthread '%s' (%s)\n", __func__, argv[arg_idx], e.what());
return 1;
}
}
if (!params.ignore_imatrix_rules && imatrix_data.empty() &&
(params.ftype == LLAMA_FTYPE_MOSTLY_IQ2_XS || params.ftype == LLAMA_FTYPE_MOSTLY_IQ2_XXS ||
params.ftype == LLAMA_FTYPE_MOSTLY_IQ2_S ||
params.ftype == LLAMA_FTYPE_MOSTLY_Q2_K_S ||
params.ftype == LLAMA_FTYPE_MOSTLY_IQ1_S ||
params.ftype == LLAMA_FTYPE_MOSTLY_IQ1_M)) {
fprintf(stderr, "\n==========================================================================================================\n");
fprintf(stderr, "Please do not use IQ1_S, IQ1_M, IQ2_S, IQ2_XXS, IQ2_XS or Q2_K_S quantization without an importance matrix\n");
fprintf(stderr, "==========================================================================================================\n\n\n");
return 1;
}
print_build_info();
fprintf(stderr, "%s: quantizing '%s' to '%s' as %s", __func__, fname_inp.c_str(), fname_out.c_str(), ftype_str.c_str());
if (params.nthread > 0) {
fprintf(stderr, " using %d threads", params.nthread);
}
fprintf(stderr, "\n");
const int64_t t_main_start_us = llama_time_us();
int64_t t_quantize_us = 0;
// load the model
{
const int64_t t_start_us = llama_time_us();
if (llama_model_quantize(fname_inp.c_str(), fname_out.c_str(), &params)) {
fprintf(stderr, "%s: failed to quantize model from '%s'\n", __func__, fname_inp.c_str());
return 1;
}
t_quantize_us = llama_time_us() - t_start_us;
}
// report timing
{
const int64_t t_main_end_us = llama_time_us();
printf("\n");
printf("%s: quantize time = %8.2f ms\n", __func__, t_quantize_us/1000.0);
printf("%s: total time = %8.2f ms\n", __func__, (t_main_end_us - t_main_start_us)/1000.0);
}
llama_backend_free();
return 0;
}
Reference in New Issue View Git Blame Copy Permalink