CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-04 09:32:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Move llama_context setup + perplexity back to main.cpp

Browse Source 
Signed-off-by: Thiago Padilha <thiago@padilha.cc>
This commit is contained in:
Thiago Padilha
2023-03-22 09:39:25 -03:00
parent d7d53b84db
commit b7f1fa6d8c
3 changed files with 128 additions and 123 deletions
Download Patch File Download Diff File Expand all files Collapse all files

124
main.cpp
View File

@@ -1,5 +1,127 @@
#include "run.h"
#include "ggml.h"
std::vector<double> softmax(const std::vector<float>& logits) {
std::vector<double> probs(logits.size());
float max_logit = logits[0];
for (float v : logits) max_logit = std::max(max_logit, v);
double sum_exp = 0.0;
for (size_t i = 0; i < logits.size(); i++) {
// Subtract the maximum logit value from the current logit value for numerical stability
float logit = logits[i] - max_logit;
double exp_logit = std::exp(logit);
sum_exp += exp_logit;
probs[i] = exp_logit;
}
for (size_t i = 0; i < probs.size(); i++) probs[i] /= sum_exp;
return probs;
}
void perplexity(llama_context * ctx, const gpt_params & params) {
// Download: https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip?ref=salesforce-research
// Run `./main --perplexity -m models/7B/ggml-model-q4_0.bin -f wiki.test.raw`
// Output: `perplexity: 13.5106 [114/114]`
auto tokens = ::llama_tokenize(ctx, params.prompt, true);
int count = 0;
double nll = 0.0;
int seq_count = tokens.size() / params.n_ctx;
fprintf(stderr, "%s : calculating perplexity over %d chunks\n", __func__, seq_count);
for (int i = 0; i < seq_count; ++i) {
int start = i * params.n_ctx;
int end = start + params.n_ctx - 1;
std::vector<llama_token> embd(tokens.begin() + start, tokens.begin() + end);
auto start_t = std::chrono::high_resolution_clock::now();
if (llama_eval(ctx, embd.data(), embd.size(), 0, params.n_threads)) {
fprintf(stderr, "%s : failed to eval\n", __func__);
return;
}
auto end_t = std::chrono::high_resolution_clock::now();
if (i == 0) {
double seconds = std::chrono::duration<double>(end_t - start_t).count();
printf("%.2f seconds per pass - ETA %.2f hours\n", seconds, (seconds * seq_count) / (60.0*60.0));
}
// We get the logits for all the tokens in the context window (params.n_ctx)
// from llama_eval above. Now, based on https://huggingface.co/docs/transformers/perplexity,
// calculate the perplexity over the last half the window (so the model always has
// some context to predict the token).
//
// We rely on the fact that attention in the forward pass only looks at previous
// tokens here, so the logits returned for each token are an accurate representation
// of what the model would have predicted at that point.
//
// Example, we have a context window of 512, we will compute perplexity for each of the
// last 256 tokens. Then, we split the input up into context window size chunks to
// process the entire prompt.
auto logits = llama_get_logits(ctx);
for (int j = params.n_ctx / 2; j < params.n_ctx - 1; ++j) {
// Calculate probability of next token, given the previous ones.
int n_vocab = llama_n_vocab(ctx);
std::vector<float> tok_logits(
logits + j * n_vocab,
logits + (j + 1) * n_vocab);
double prob = softmax(tok_logits)[tokens[start + j + 1]];
nll += -std::log(prob);
++count;
}
// perplexity is e^(average negative log-likelihood)
printf("[%d]%.4lf,", i + 1, std::exp(nll / count));
fflush(stdout);
}
printf("\n");
}
int main(int argc, char ** argv) {
return run(argc, argv);
// has to be called once at the start of the program to init ggml stuff
ggml_time_init();
gpt_params params;
params.model = "models/llama-7B/ggml-model.bin";
if (gpt_params_parse(argc, argv, params) == false) {
return 1;
}
if (params.n_ctx > 2048) {
fprintf(stderr, "%s: warning: model does not support context sizes greater than 2048 tokens (%d specified);"
"expect poor results\n", __func__, params.n_ctx);
}
llama_context * ctx;
// load the model
{
auto lparams = llama_context_default_params();
lparams.n_ctx = params.n_ctx;
lparams.n_parts = params.n_parts;
lparams.seed = params.seed;
lparams.f16_kv = params.memory_f16;
lparams.logits_all = params.perplexity;
ctx = llama_init_from_file(params.model.c_str(), lparams);
if (ctx == NULL) {
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
return 1;
}
}
// print system information
{
fprintf(stderr, "\n");
fprintf(stderr, "system_info: n_threads = %d / %d | %s\n",
params.n_threads, std::thread::hardware_concurrency(), llama_print_system_info());
}
if (params.perplexity) {
perplexity(ctx, params);
exit(0);
}
return run(ctx, params);
}

122
run.cpp
View File

@@ -1,5 +1,4 @@
#include "utils.h"
#include "ggml.h"
#include "llama.h"
#include <cassert>
@@ -65,79 +64,6 @@ void set_console_state(console_state new_st)
}
}
std::vector<double> softmax(const std::vector<float>& logits) {
std::vector<double> probs(logits.size());
float max_logit = logits[0];
for (float v : logits) max_logit = std::max(max_logit, v);
double sum_exp = 0.0;
for (size_t i = 0; i < logits.size(); i++) {
// Subtract the maximum logit value from the current logit value for numerical stability
float logit = logits[i] - max_logit;
double exp_logit = std::exp(logit);
sum_exp += exp_logit;
probs[i] = exp_logit;
}
for (size_t i = 0; i < probs.size(); i++) probs[i] /= sum_exp;
return probs;
}
void perplexity(llama_context * ctx, const gpt_params & params) {
// Download: https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip?ref=salesforce-research
// Run `./main --perplexity -m models/7B/ggml-model-q4_0.bin -f wiki.test.raw`
// Output: `perplexity: 13.5106 [114/114]`
auto tokens = ::llama_tokenize(ctx, params.prompt, true);
int count = 0;
double nll = 0.0;
int seq_count = tokens.size() / params.n_ctx;
fprintf(stderr, "%s : calculating perplexity over %d chunks\n", __func__, seq_count);
for (int i = 0; i < seq_count; ++i) {
int start = i * params.n_ctx;
int end = start + params.n_ctx - 1;
std::vector<llama_token> embd(tokens.begin() + start, tokens.begin() + end);
auto start_t = std::chrono::high_resolution_clock::now();
if (llama_eval(ctx, embd.data(), embd.size(), 0, params.n_threads)) {
fprintf(stderr, "%s : failed to eval\n", __func__);
return;
}
auto end_t = std::chrono::high_resolution_clock::now();
if (i == 0) {
double seconds = std::chrono::duration<double>(end_t - start_t).count();
printf("%.2f seconds per pass - ETA %.2f hours\n", seconds, (seconds * seq_count) / (60.0*60.0));
}
// We get the logits for all the tokens in the context window (params.n_ctx)
// from llama_eval above. Now, based on https://huggingface.co/docs/transformers/perplexity,
// calculate the perplexity over the last half the window (so the model always has
// some context to predict the token).
//
// We rely on the fact that attention in the forward pass only looks at previous
// tokens here, so the logits returned for each token are an accurate representation
// of what the model would have predicted at that point.
//
// Example, we have a context window of 512, we will compute perplexity for each of the
// last 256 tokens. Then, we split the input up into context window size chunks to
// process the entire prompt.
auto logits = llama_get_logits(ctx);
for (int j = params.n_ctx / 2; j < params.n_ctx - 1; ++j) {
// Calculate probability of next token, given the previous ones.
int n_vocab = llama_n_vocab(ctx);
std::vector<float> tok_logits(
logits + j * n_vocab,
logits + (j + 1) * n_vocab);
double prob = softmax(tok_logits)[tokens[start + j + 1]];
nll += -std::log(prob);
++count;
}
// perplexity is e^(average negative log-likelihood)
printf("[%d]%.4lf,", i + 1, std::exp(nll / count));
fflush(stdout);
}
printf("\n");
}
static bool is_interacting = false;
#if defined (__unix__) || (defined (__APPLE__) && defined (__MACH__)) || defined (_WIN32)
@@ -154,21 +80,7 @@ void sigint_handler(int signo) {
}
#endif
int run(int argc, char ** argv) {
// has to be called once at the start of the program to init ggml stuff
ggml_time_init();
gpt_params params;
params.model = "models/llama-7B/ggml-model.bin";
if (gpt_params_parse(argc, argv, params) == false) {
return 1;
}
if (params.n_ctx > 2048) {
fprintf(stderr, "%s: warning: model does not support context sizes greater than 2048 tokens (%d specified);"
"expect poor results\n", __func__, params.n_ctx);
}
int run(llama_context * ctx, gpt_params params) {
if (params.seed <= 0) {
params.seed = time(NULL);
@@ -188,33 +100,6 @@ int run(int argc, char ** argv) {
// params.prompt = R"(// this function checks if the number n is prime
//bool is_prime(int n) {)";
llama_context * ctx;
// load the model
{
auto lparams = llama_context_default_params();
lparams.n_ctx = params.n_ctx;
lparams.n_parts = params.n_parts;
lparams.seed = params.seed;
lparams.f16_kv = params.memory_f16;
lparams.logits_all = params.perplexity;
ctx = llama_init_from_file(params.model.c_str(), lparams);
if (ctx == NULL) {
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
return 1;
}
}
// print system information
{
fprintf(stderr, "\n");
fprintf(stderr, "system_info: n_threads = %d / %d | %s\n",
params.n_threads, std::thread::hardware_concurrency(), llama_print_system_info());
}
// determine the required inference memory per token:
// TODO: better way to do that
{
@@ -222,11 +107,6 @@ int run(int argc, char ** argv) {
llama_eval(ctx, tmp.data(), tmp.size(), 0, params.n_threads);
}
if (params.perplexity) {
perplexity(ctx, params);
exit(0);
}
int n_past = 0;
// Add a space in front of the first character to match OG llama tokenizer behavior

5
run.h
View File

@@ -1,3 +1,6 @@
#pragma once
int run(int argc, char ** argv);
#include "llama.h"
#include "utils.h"
int run(llama_context * ctx, gpt_params params);