2023-04-15 19:57:32 +00:00
|
|
|
#include "utils.h"
|
|
|
|
|
|
|
|
#include <fstream>
|
|
|
|
#include <regex>
|
2023-05-30 16:59:00 +00:00
|
|
|
|
|
|
|
void replace(std::string & str, const std::string & needle, const std::string & replacement) {
|
|
|
|
size_t pos = 0;
|
|
|
|
while ((pos = str.find(needle, pos)) != std::string::npos) {
|
|
|
|
str.replace(pos, needle.length(), replacement);
|
|
|
|
pos += replacement.length();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
std::map<std::string, int32_t> json_parse(const std::string & fname) {
|
|
|
|
std::map<std::string, int32_t> result;
|
|
|
|
|
|
|
|
// read file into string
|
|
|
|
std::string json;
|
|
|
|
{
|
|
|
|
std::ifstream ifs(fname);
|
|
|
|
if (!ifs) {
|
|
|
|
fprintf(stderr, "Failed to open %s\n", fname.c_str());
|
|
|
|
exit(1);
|
|
|
|
}
|
|
|
|
|
|
|
|
json = std::string((std::istreambuf_iterator<char>(ifs)),
|
|
|
|
(std::istreambuf_iterator<char>()));
|
|
|
|
}
|
|
|
|
|
|
|
|
if (json[0] != '{') {
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
// parse json
|
|
|
|
{
|
|
|
|
bool has_key = false;
|
|
|
|
bool in_token = false;
|
|
|
|
|
|
|
|
std::string str_key = "";
|
|
|
|
std::string str_val = "";
|
|
|
|
|
|
|
|
int n = json.size();
|
|
|
|
for (int i = 1; i < n; ++i) {
|
|
|
|
if (!in_token) {
|
|
|
|
if (json[i] == ' ') continue;
|
|
|
|
if (json[i] == '"') {
|
|
|
|
in_token = true;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if (json[i] == '\\' && i+1 < n) {
|
|
|
|
if (has_key == false) {
|
|
|
|
str_key += json[i];
|
|
|
|
} else {
|
|
|
|
str_val += json[i];
|
|
|
|
}
|
|
|
|
++i;
|
|
|
|
} else if (json[i] == '"') {
|
|
|
|
if (has_key == false) {
|
|
|
|
has_key = true;
|
|
|
|
++i;
|
|
|
|
while (json[i] == ' ') ++i;
|
|
|
|
++i; // :
|
|
|
|
while (json[i] == ' ') ++i;
|
|
|
|
if (json[i] != '\"') {
|
|
|
|
while (json[i] != ',' && json[i] != '}') {
|
|
|
|
str_val += json[i++];
|
|
|
|
}
|
|
|
|
has_key = false;
|
|
|
|
} else {
|
|
|
|
in_token = true;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
has_key = false;
|
|
|
|
}
|
|
|
|
|
|
|
|
::replace(str_key, "\\u0120", " " ); // \u0120 -> space
|
|
|
|
::replace(str_key, "\\u010a", "\n"); // \u010a -> new line
|
|
|
|
::replace(str_key, "\\\"", "\""); // \\\" -> "
|
|
|
|
|
|
|
|
try {
|
|
|
|
result[str_key] = std::stoi(str_val);
|
|
|
|
} catch (...) {
|
|
|
|
//fprintf(stderr, "%s: ignoring key '%s' with value '%s'\n", fname.c_str(), str_key.c_str(), str_val.c_str());
|
|
|
|
|
|
|
|
}
|
|
|
|
str_key = "";
|
|
|
|
str_val = "";
|
|
|
|
in_token = false;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
if (has_key == false) {
|
|
|
|
str_key += json[i];
|
|
|
|
} else {
|
|
|
|
str_val += json[i];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
std::vector<gpt_vocab::id> gpt_tokenize_inner(const gpt_vocab & vocab, const std::string & text) {
|
|
|
|
std::vector<std::string> words;
|
|
|
|
|
|
|
|
// first split the text into words
|
|
|
|
{
|
|
|
|
std::string str = text;
|
|
|
|
std::string pat = R"('s|'t|'re|'ve|'m|'ll|'d| ?[[:alpha:]]+| ?[[:digit:]]+| ?[^\s[:alpha:][:digit:]]+|\s+(?!\S)|\s+)";
|
|
|
|
|
|
|
|
std::regex re(pat);
|
|
|
|
std::smatch m;
|
|
|
|
|
|
|
|
while (std::regex_search(str, m, re)) {
|
|
|
|
for (auto x : m) {
|
|
|
|
words.push_back(x);
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
str = m.suffix();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// find the longest tokens that form the words:
|
|
|
|
std::vector<gpt_vocab::id> tokens;
|
|
|
|
for (const auto & word : words) {
|
|
|
|
if (word.size() == 0) continue;
|
|
|
|
|
|
|
|
int i = 0;
|
|
|
|
int n = word.size();
|
|
|
|
while (i < n) {
|
|
|
|
int j = n;
|
|
|
|
while (j > i) {
|
|
|
|
auto it = vocab.token_to_id.find(word.substr(i, j-i));
|
|
|
|
if (it != vocab.token_to_id.end()) {
|
|
|
|
tokens.push_back(it->second);
|
|
|
|
i = j;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
--j;
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
if (i == n) {
|
|
|
|
break;
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
if (j == i) {
|
|
|
|
auto sub = word.substr(i, 1);
|
|
|
|
if (vocab.token_to_id.find(sub) != vocab.token_to_id.end()) {
|
|
|
|
tokens.push_back(vocab.token_to_id.at(sub));
|
|
|
|
} else {
|
|
|
|
fprintf(stderr, "%s: unknown token '%s'\n", __func__, sub.data());
|
|
|
|
}
|
|
|
|
++i;
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return tokens;
|
|
|
|
}
|
|
|
|
|
|
|
|
std::string regex_escape(const std::string &s) {
|
|
|
|
static const std::regex metacharacters(R"([\.\^\$\-\+\(\)\[\]\{\}\|\?\*])");
|
|
|
|
return std::regex_replace(s, metacharacters, "\\$&");
|
|
|
|
}
|
|
|
|
|
|
|
|
std::vector<gpt_vocab::id> gpt_tokenize(const gpt_vocab & vocab, const std::string & text) {
|
|
|
|
// Generate the subpattern from the special_tokens vector if it's not empty
|
|
|
|
if (!vocab.special_tokens.empty()) {
|
|
|
|
std::vector<gpt_vocab::id> out;
|
|
|
|
std::vector<std::string> chunks;
|
|
|
|
std::string str = text;
|
|
|
|
std::string special_tokens_subpattern;
|
|
|
|
for (const auto &token : vocab.special_tokens) {
|
|
|
|
if (!special_tokens_subpattern.empty()) {
|
|
|
|
special_tokens_subpattern += "|";
|
2023-05-16 00:30:45 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
special_tokens_subpattern += regex_escape(token);
|
|
|
|
}
|
|
|
|
std::regex re(special_tokens_subpattern);
|
|
|
|
std::smatch m;
|
|
|
|
while (std::regex_search(str, m, re)) {
|
|
|
|
auto tok = vocab.token_to_id.find(m.str());
|
|
|
|
if (tok != vocab.token_to_id.end()) {
|
|
|
|
auto tokid = tok->second;
|
|
|
|
auto pfxtoks = gpt_tokenize_inner(vocab, m.prefix());
|
|
|
|
out.insert(out.end(), pfxtoks.begin(), pfxtoks.end());
|
|
|
|
out.push_back(tokid);
|
|
|
|
str = m.suffix();
|
2023-05-16 00:30:45 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
}
|
|
|
|
if (!str.empty()) {
|
|
|
|
auto tokrest = gpt_tokenize_inner(vocab, str);
|
|
|
|
out.insert(out.end(), tokrest.begin(), tokrest.end());
|
|
|
|
}
|
|
|
|
return out;
|
|
|
|
} else {
|
|
|
|
return gpt_tokenize_inner(vocab, text);
|
2023-05-16 00:30:45 +00:00
|
|
|
}
|
2023-05-30 16:59:00 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
bool gpt_vocab_init(const std::string & fname, gpt_vocab & vocab) {
|
|
|
|
printf("%s: loading vocab from '%s'\n", __func__, fname.c_str());
|
|
|
|
|
|
|
|
vocab.token_to_id = ::json_parse(fname);
|
|
|
|
|
|
|
|
for (const auto & kv : vocab.token_to_id) {
|
|
|
|
vocab.id_to_token[kv.second] = kv.first;
|
|
|
|
}
|
|
|
|
|
|
|
|
printf("%s: vocab size = %d\n", __func__, (int) vocab.token_to_id.size());
|
|
|
|
|
|
|
|
// print the vocabulary
|
|
|
|
//for (auto kv : vocab.token_to_id) {
|
|
|
|
// printf("'%s' -> %d\n", kv.first.data(), kv.second);
|
|
|
|
//}
|
|
|
|
|
|
|
|
return true;
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
gpt_vocab::id gpt_sample_top_k_top_p(
|
2023-05-16 00:42:20 +00:00
|
|
|
const size_t actualVocabSize,
|
2023-04-25 12:38:29 +00:00
|
|
|
const int32_t * last_n_tokens_data,
|
|
|
|
int last_n_tokens_size,
|
|
|
|
const std::vector<float> logits,
|
2023-04-15 19:57:32 +00:00
|
|
|
int top_k,
|
|
|
|
double top_p,
|
|
|
|
double temp,
|
2023-04-25 12:38:29 +00:00
|
|
|
float repeat_penalty,
|
2023-04-15 19:57:32 +00:00
|
|
|
std::mt19937 & rng) {
|
2023-05-16 00:42:20 +00:00
|
|
|
int n_logits = actualVocabSize;
|
2023-04-15 19:57:32 +00:00
|
|
|
|
2023-04-25 12:38:29 +00:00
|
|
|
const auto last_n_tokens = std::vector<int32_t>(last_n_tokens_data, last_n_tokens_data + last_n_tokens_size);
|
2023-06-08 18:08:10 +00:00
|
|
|
const auto * plogits = logits.data();
|
2023-04-25 12:38:29 +00:00
|
|
|
|
2023-06-08 18:08:30 +00:00
|
|
|
if (temp <= 0) {
|
|
|
|
// select the token with the highest logit directly
|
|
|
|
float max_logit = plogits[0];
|
|
|
|
gpt_vocab::id max_id = 0;
|
|
|
|
|
|
|
|
for (int i = 1; i < n_logits; ++i) {
|
|
|
|
if (plogits[i] > max_logit) {
|
|
|
|
max_logit = plogits[i];
|
|
|
|
max_id = i;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return max_id;
|
|
|
|
}
|
2023-04-15 19:57:32 +00:00
|
|
|
std::vector<std::pair<double, gpt_vocab::id>> logits_id;
|
|
|
|
logits_id.reserve(n_logits);
|
|
|
|
|
|
|
|
{
|
2023-04-25 12:38:29 +00:00
|
|
|
const float scale = 1.0f/temp;
|
2023-04-15 19:57:32 +00:00
|
|
|
for (int i = 0; i < n_logits; ++i) {
|
2023-04-25 12:38:29 +00:00
|
|
|
// repetition penalty from ctrl paper (https://arxiv.org/abs/1909.05858)
|
|
|
|
// credit https://github.com/facebookresearch/llama/compare/main...shawwn:llama:main
|
|
|
|
if (std::find(last_n_tokens.begin(), last_n_tokens.end(), i) != last_n_tokens.end()) {
|
|
|
|
// if score < 0 then repetition penalty has to multiplied to reduce the previous token probability
|
|
|
|
if (plogits[i] < 0.0f) {
|
|
|
|
logits_id.push_back(std::make_pair(plogits[i]*scale*repeat_penalty, i));
|
|
|
|
} else {
|
|
|
|
logits_id.push_back(std::make_pair(plogits[i]*scale/repeat_penalty, i));
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
logits_id.push_back(std::make_pair(plogits[i]*scale, i));
|
|
|
|
}
|
2023-04-15 19:57:32 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// find the top K tokens
|
|
|
|
std::partial_sort(
|
|
|
|
logits_id.begin(),
|
|
|
|
logits_id.begin() + top_k, logits_id.end(),
|
|
|
|
[](const std::pair<double, gpt_vocab::id> & a, const std::pair<double, gpt_vocab::id> & b) {
|
|
|
|
return a.first > b.first;
|
|
|
|
});
|
|
|
|
|
|
|
|
logits_id.resize(top_k);
|
|
|
|
|
|
|
|
double maxl = -INFINITY;
|
|
|
|
for (const auto & kv : logits_id) {
|
|
|
|
maxl = std::max(maxl, kv.first);
|
|
|
|
}
|
|
|
|
|
|
|
|
// compute probs for the top K tokens
|
|
|
|
std::vector<double> probs;
|
|
|
|
probs.reserve(logits_id.size());
|
|
|
|
|
|
|
|
double sum = 0.0;
|
|
|
|
for (const auto & kv : logits_id) {
|
|
|
|
double p = exp(kv.first - maxl);
|
|
|
|
probs.push_back(p);
|
|
|
|
sum += p;
|
|
|
|
}
|
|
|
|
|
|
|
|
// normalize the probs
|
|
|
|
for (auto & p : probs) {
|
|
|
|
p /= sum;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (top_p < 1.0f) {
|
|
|
|
double cumsum = 0.0f;
|
|
|
|
for (int i = 0; i < top_k; i++) {
|
|
|
|
cumsum += probs[i];
|
|
|
|
if (cumsum >= top_p) {
|
|
|
|
top_k = i + 1;
|
|
|
|
probs.resize(top_k);
|
|
|
|
logits_id.resize(top_k);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
cumsum = 1.0/cumsum;
|
|
|
|
for (int i = 0; i < (int) probs.size(); i++) {
|
|
|
|
probs[i] *= cumsum;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
//printf("\n");
|
|
|
|
//for (int i = 0; i < (int) probs.size(); i++) {
|
|
|
|
// printf("%d: '%s' %f\n", i, vocab.id_to_token.at(logits_id[i].second).c_str(), probs[i]);
|
|
|
|
//}
|
|
|
|
//exit(0);
|
|
|
|
|
|
|
|
std::discrete_distribution<> dist(probs.begin(), probs.end());
|
|
|
|
int idx = dist(rng);
|
|
|
|
|
|
|
|
return logits_id[idx].second;
|
2023-05-30 16:59:00 +00:00
|
|
|
}
|