Fix editorconfig
This commit is contained in:
Alisamar Husain
parent
2b50d21423
commit
d2af46e371
3 changed files with 8874 additions and 8969 deletions
File diff suppressed because it is too large
Load diff
|
|
@ -32,13 +32,18 @@ static llama_context ** g_ctx;
|
|||
static bool is_interacting = false;
|
||||
|
||||
#if defined(__unix__) || (defined(__APPLE__) && defined(__MACH__)) || defined(_WIN32)
|
||||
void sigint_handler(int signo) {
|
||||
void sigint_handler(int signo)
|
||||
{
|
||||
set_console_color(con_st, CONSOLE_COLOR_DEFAULT);
|
||||
printf("\n"); // this also force flush stdout.
|
||||
if (signo == SIGINT) {
|
||||
if (!is_interacting) {
|
||||
if (signo == SIGINT)
|
||||
{
|
||||
if (!is_interacting)
|
||||
{
|
||||
is_interacting = true;
|
||||
} else {
|
||||
}
|
||||
else
|
||||
{
|
||||
llama_print_timings(*g_ctx);
|
||||
_exit(130);
|
||||
}
|
||||
|
|
@ -46,16 +51,18 @@ void sigint_handler(int signo) {
|
|||
}
|
||||
#endif
|
||||
|
||||
|
||||
auto const BINDPORT = 8001;
|
||||
|
||||
int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
||||
if (!params.lora_adapter.empty()) {
|
||||
int run_llama(llama_context *ctx, gpt_params params, std::ostream *outfile)
|
||||
{
|
||||
if (!params.lora_adapter.empty())
|
||||
{
|
||||
int err = llama_apply_lora_from_file(ctx,
|
||||
params.lora_adapter.c_str(),
|
||||
params.lora_base.empty() ? NULL : params.lora_base.c_str(),
|
||||
params.n_threads);
|
||||
if (err != 0) {
|
||||
if (err != 0)
|
||||
{
|
||||
fprintf(stderr, "%s: error: failed to apply lora adapter\n", __func__);
|
||||
return 1;
|
||||
}
|
||||
|
|
@ -70,14 +77,17 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
// determine the maximum memory usage needed to do inference for the given n_batch and n_predict parameters
|
||||
// uncomment the "used_mem" line in llama.cpp to see the results
|
||||
if (params.mem_test) {
|
||||
if (params.mem_test)
|
||||
{
|
||||
{
|
||||
const std::vector<llama_token> tmp(params.n_batch, 0);
|
||||
llama_eval(ctx, tmp.data(), tmp.size(), 0, params.n_threads);
|
||||
}
|
||||
|
||||
{
|
||||
const std::vector<llama_token> tmp = { 0, };
|
||||
const std::vector<llama_token> tmp = {
|
||||
0,
|
||||
};
|
||||
llama_eval(ctx, tmp.data(), tmp.size(), params.n_predict - 1, params.n_threads);
|
||||
}
|
||||
|
||||
|
|
@ -95,13 +105,15 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
const int n_ctx = llama_n_ctx(ctx);
|
||||
|
||||
if ((int) embd_inp.size() > n_ctx - 4) {
|
||||
if ((int)embd_inp.size() > n_ctx - 4)
|
||||
{
|
||||
fprintf(stderr, "%s: error: prompt is too long (%d tokens, max %d)\n", __func__, (int)embd_inp.size(), n_ctx - 4);
|
||||
return 1;
|
||||
}
|
||||
|
||||
// number of tokens to keep when resetting context
|
||||
if (params.n_keep < 0 || params.n_keep > (int)embd_inp.size() || params.instruct) {
|
||||
if (params.n_keep < 0 || params.n_keep > (int)embd_inp.size() || params.instruct)
|
||||
{
|
||||
params.n_keep = (int)embd_inp.size();
|
||||
}
|
||||
|
||||
|
|
@ -110,29 +122,35 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
const auto inp_sfx = ::llama_tokenize(ctx, "\n\n### Response:\n\n", false);
|
||||
|
||||
// in instruct mode, we inject a prefix and a suffix to each input by the user
|
||||
if (params.instruct) {
|
||||
if (params.instruct)
|
||||
{
|
||||
params.interactive_first = true;
|
||||
params.antiprompt.push_back("### Instruction:\n\n");
|
||||
}
|
||||
|
||||
// enable interactive mode if reverse prompt or interactive start is specified
|
||||
if (params.antiprompt.size() != 0 || params.interactive_first) {
|
||||
if (params.antiprompt.size() != 0 || params.interactive_first)
|
||||
{
|
||||
params.interactive = true;
|
||||
}
|
||||
|
||||
// determine newline token
|
||||
auto llama_token_newline = ::llama_tokenize(ctx, "\n", false);
|
||||
|
||||
if (params.verbose_prompt) {
|
||||
if (params.verbose_prompt)
|
||||
{
|
||||
fprintf(stderr, "\n");
|
||||
fprintf(stderr, "%s: prompt: '%s'\n", __func__, params.prompt.c_str());
|
||||
fprintf(stderr, "%s: number of tokens in prompt = %zu\n", __func__, embd_inp.size());
|
||||
for (int i = 0; i < (int) embd_inp.size(); i++) {
|
||||
for (int i = 0; i < (int)embd_inp.size(); i++)
|
||||
{
|
||||
fprintf(stderr, "%6d -> '%s'\n", embd_inp[i], llama_token_to_str(ctx, embd_inp[i]));
|
||||
}
|
||||
if (params.n_keep > 0) {
|
||||
if (params.n_keep > 0)
|
||||
{
|
||||
fprintf(stderr, "%s: static prompt based on n_keep: '", __func__);
|
||||
for (int i = 0; i < params.n_keep; i++) {
|
||||
for (int i = 0; i < params.n_keep; i++)
|
||||
{
|
||||
fprintf(stderr, "%s", llama_token_to_str(ctx, embd_inp[i]));
|
||||
}
|
||||
fprintf(stderr, "'\n");
|
||||
|
|
@ -140,7 +158,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
fprintf(stderr, "\n");
|
||||
}
|
||||
|
||||
if (params.interactive) {
|
||||
if (params.interactive)
|
||||
{
|
||||
#if defined(__unix__) || (defined(__APPLE__) && defined(__MACH__))
|
||||
struct sigaction sigint_action;
|
||||
sigint_action.sa_handler = sigint_handler;
|
||||
|
|
@ -153,13 +172,16 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
fprintf(stderr, "%s: interactive mode on.\n", __func__);
|
||||
|
||||
if (params.antiprompt.size()) {
|
||||
for (auto antiprompt : params.antiprompt) {
|
||||
if (params.antiprompt.size())
|
||||
{
|
||||
for (auto antiprompt : params.antiprompt)
|
||||
{
|
||||
fprintf(stderr, "Reverse prompt: '%s'\n", antiprompt.c_str());
|
||||
}
|
||||
}
|
||||
|
||||
if (!params.input_prefix.empty()) {
|
||||
if (!params.input_prefix.empty())
|
||||
{
|
||||
fprintf(stderr, "Input prefix: '%s'\n", params.input_prefix.c_str());
|
||||
}
|
||||
}
|
||||
|
|
@ -172,7 +194,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
std::vector<llama_token> last_n_tokens(n_ctx);
|
||||
std::fill(last_n_tokens.begin(), last_n_tokens.end(), 0);
|
||||
|
||||
if (params.interactive) {
|
||||
if (params.interactive)
|
||||
{
|
||||
fprintf(stderr, "== Running in interactive mode. ==\n"
|
||||
#if defined(__unix__) || (defined(__APPLE__) && defined(__MACH__)) || defined(_WIN32)
|
||||
" - Press Ctrl+C to interject at any time.\n"
|
||||
|
|
@ -194,14 +217,17 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
std::vector<llama_token> embd;
|
||||
|
||||
while (n_remain != 0 || params.interactive) {
|
||||
while (n_remain != 0 || params.interactive)
|
||||
{
|
||||
// predict
|
||||
if (embd.size() > 0) {
|
||||
if (embd.size() > 0)
|
||||
{
|
||||
// infinite text generation via context swapping
|
||||
// if we run out of context:
|
||||
// - take the n_keep first tokens from the original prompt (via n_past)
|
||||
// - take half of the last (n_ctx - n_keep) tokens and recompute the logits in batches
|
||||
if (n_past + (int) embd.size() > n_ctx) {
|
||||
if (n_past + (int)embd.size() > n_ctx)
|
||||
{
|
||||
const int n_left = n_past - params.n_keep;
|
||||
|
||||
n_past = params.n_keep;
|
||||
|
|
@ -220,12 +246,15 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
// evaluate tokens in batches
|
||||
// embd is typically prepared beforehand to fit within a batch, but not always
|
||||
for (int i = 0; i < (int) embd.size(); i += params.n_batch) {
|
||||
for (int i = 0; i < (int)embd.size(); i += params.n_batch)
|
||||
{
|
||||
int n_eval = (int)embd.size() - i;
|
||||
if (n_eval > params.n_batch) {
|
||||
if (n_eval > params.n_batch)
|
||||
{
|
||||
n_eval = params.n_batch;
|
||||
}
|
||||
if (llama_eval(ctx, &embd[i], n_eval, n_past, params.n_threads)) {
|
||||
if (llama_eval(ctx, &embd[i], n_eval, n_past, params.n_threads))
|
||||
{
|
||||
fprintf(stderr, "%s : failed to eval\n", __func__);
|
||||
return 1;
|
||||
}
|
||||
|
|
@ -235,7 +264,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
embd.clear();
|
||||
|
||||
if ((int) embd_inp.size() <= n_consumed && !is_interacting) {
|
||||
if ((int)embd_inp.size() <= n_consumed && !is_interacting)
|
||||
{
|
||||
// out of user input, sample next token
|
||||
const int32_t top_k = params.top_k;
|
||||
const float top_p = params.top_p;
|
||||
|
|
@ -247,7 +277,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
{
|
||||
auto logits = llama_get_logits(ctx);
|
||||
|
||||
if (params.ignore_eos) {
|
||||
if (params.ignore_eos)
|
||||
{
|
||||
logits[llama_token_eos()] = 0;
|
||||
}
|
||||
|
||||
|
|
@ -260,9 +291,11 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
}
|
||||
|
||||
// replace end of text token with newline token when in interactive mode
|
||||
if (id == llama_token_eos() && params.interactive && !params.instruct) {
|
||||
if (id == llama_token_eos() && params.interactive && !params.instruct)
|
||||
{
|
||||
id = llama_token_newline.front();
|
||||
if (params.antiprompt.size() != 0) {
|
||||
if (params.antiprompt.size() != 0)
|
||||
{
|
||||
// tokenize and inject first reverse prompt
|
||||
const auto first_antiprompt = ::llama_tokenize(ctx, params.antiprompt.front(), false);
|
||||
embd_inp.insert(embd_inp.end(), first_antiprompt.begin(), first_antiprompt.end());
|
||||
|
|
@ -277,45 +310,57 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
// decrement remaining sampling budget
|
||||
--n_remain;
|
||||
} else {
|
||||
}
|
||||
else
|
||||
{
|
||||
// some user input remains from prompt or interaction, forward it to processing
|
||||
while ((int) embd_inp.size() > n_consumed) {
|
||||
while ((int)embd_inp.size() > n_consumed)
|
||||
{
|
||||
embd.push_back(embd_inp[n_consumed]);
|
||||
last_n_tokens.erase(last_n_tokens.begin());
|
||||
last_n_tokens.push_back(embd_inp[n_consumed]);
|
||||
++n_consumed;
|
||||
if ((int) embd.size() >= params.n_batch) {
|
||||
if ((int)embd.size() >= params.n_batch)
|
||||
{
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// display text
|
||||
if (!input_noecho) {
|
||||
for (auto id : embd) {
|
||||
if (!input_noecho)
|
||||
{
|
||||
for (auto id : embd)
|
||||
{
|
||||
*outfile << llama_token_to_str(ctx, id) << std::flush;
|
||||
}
|
||||
}
|
||||
// reset color to default if we there is no pending user input
|
||||
if (!input_noecho && (int)embd_inp.size() == n_consumed) {
|
||||
if (!input_noecho && (int)embd_inp.size() == n_consumed)
|
||||
{
|
||||
set_console_color(con_st, CONSOLE_COLOR_DEFAULT);
|
||||
}
|
||||
|
||||
// in interactive mode, and not currently processing queued inputs;
|
||||
// check if we should prompt the user for more
|
||||
if (params.interactive && (int) embd_inp.size() <= n_consumed) {
|
||||
if (params.interactive && (int)embd_inp.size() <= n_consumed)
|
||||
{
|
||||
|
||||
// check for reverse prompt
|
||||
if (params.antiprompt.size()) {
|
||||
if (params.antiprompt.size())
|
||||
{
|
||||
std::string last_output;
|
||||
for (auto id : last_n_tokens) {
|
||||
for (auto id : last_n_tokens)
|
||||
{
|
||||
last_output += llama_token_to_str(ctx, id);
|
||||
}
|
||||
|
||||
is_antiprompt = false;
|
||||
// Check if each of the reverse prompts appears at the end of the output.
|
||||
for (std::string & antiprompt : params.antiprompt) {
|
||||
if (last_output.find(antiprompt.c_str(), last_output.length() - antiprompt.length(), antiprompt.length()) != std::string::npos) {
|
||||
for (std::string &antiprompt : params.antiprompt)
|
||||
{
|
||||
if (last_output.find(antiprompt.c_str(), last_output.length() - antiprompt.length(), antiprompt.length()) != std::string::npos)
|
||||
{
|
||||
is_interacting = true;
|
||||
is_antiprompt = true;
|
||||
set_console_color(con_st, CONSOLE_COLOR_USER_INPUT);
|
||||
|
|
@ -325,7 +370,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
}
|
||||
}
|
||||
|
||||
if (n_past > 0 && is_interacting) {
|
||||
if (n_past > 0 && is_interacting)
|
||||
{
|
||||
// potentially set color to indicate we are taking user input
|
||||
set_console_color(con_st, CONSOLE_COLOR_USER_INPUT);
|
||||
|
||||
|
|
@ -334,35 +380,43 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
signal(SIGINT, sigint_handler);
|
||||
#endif
|
||||
|
||||
if (params.instruct) {
|
||||
if (params.instruct)
|
||||
{
|
||||
printf("\n> ");
|
||||
}
|
||||
|
||||
std::string buffer;
|
||||
if (!params.input_prefix.empty()) {
|
||||
if (!params.input_prefix.empty())
|
||||
{
|
||||
buffer += params.input_prefix;
|
||||
printf("%s", buffer.c_str());
|
||||
}
|
||||
|
||||
std::string line;
|
||||
bool another_line = true;
|
||||
do {
|
||||
do
|
||||
{
|
||||
#if defined(_WIN32)
|
||||
std::wstring wline;
|
||||
if (!std::getline(std::wcin, wline)) {
|
||||
if (!std::getline(std::wcin, wline))
|
||||
{
|
||||
// input stream is bad or EOF received
|
||||
return 0;
|
||||
}
|
||||
win32_utf8_encode(wline, line);
|
||||
#else
|
||||
if (!std::getline(std::cin, line)) {
|
||||
if (!std::getline(std::cin, line))
|
||||
{
|
||||
// input stream is bad or EOF received
|
||||
return 0;
|
||||
}
|
||||
#endif
|
||||
if (line.empty() || line.back() != '\\') {
|
||||
if (line.empty() || line.back() != '\\')
|
||||
{
|
||||
another_line = false;
|
||||
} else {
|
||||
}
|
||||
else
|
||||
{
|
||||
line.pop_back(); // Remove the continue character
|
||||
}
|
||||
buffer += line + '\n'; // Append the line to the result
|
||||
|
|
@ -373,10 +427,12 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
|
||||
// Add tokens to embd only if the input buffer is non-empty
|
||||
// Entering a empty line lets the user pass control back
|
||||
if (buffer.length() > 1) {
|
||||
if (buffer.length() > 1)
|
||||
{
|
||||
|
||||
// instruct mode: insert instruction prefix
|
||||
if (params.instruct && !is_antiprompt) {
|
||||
if (params.instruct && !is_antiprompt)
|
||||
{
|
||||
n_consumed = embd_inp.size();
|
||||
embd_inp.insert(embd_inp.end(), inp_pfx.begin(), inp_pfx.end());
|
||||
}
|
||||
|
|
@ -385,7 +441,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
embd_inp.insert(embd_inp.end(), line_inp.begin(), line_inp.end());
|
||||
|
||||
// instruct mode: insert response suffix
|
||||
if (params.instruct) {
|
||||
if (params.instruct)
|
||||
{
|
||||
embd_inp.insert(embd_inp.end(), inp_sfx.begin(), inp_sfx.end());
|
||||
}
|
||||
|
||||
|
|
@ -395,23 +452,29 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
input_noecho = true; // do not echo this again
|
||||
}
|
||||
|
||||
if (n_past > 0) {
|
||||
if (n_past > 0)
|
||||
{
|
||||
is_interacting = false;
|
||||
}
|
||||
}
|
||||
|
||||
// end of text token
|
||||
if (!embd.empty() && embd.back() == llama_token_eos()) {
|
||||
if (params.instruct) {
|
||||
if (!embd.empty() && embd.back() == llama_token_eos())
|
||||
{
|
||||
if (params.instruct)
|
||||
{
|
||||
is_interacting = true;
|
||||
} else {
|
||||
}
|
||||
else
|
||||
{
|
||||
fprintf(stderr, " [end of text]\n");
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// In interactive mode, respect the maximum number of tokens and drop back to user input when reached.
|
||||
if (params.interactive && n_remain <= 0 && params.n_predict != -1) {
|
||||
if (params.interactive && n_remain <= 0 && params.n_predict != -1)
|
||||
{
|
||||
n_remain = params.n_predict;
|
||||
is_interacting = true;
|
||||
}
|
||||
|
|
@ -429,7 +492,8 @@ int run_llama(llama_context * ctx, gpt_params params, std::ostream * outfile) {
|
|||
return 0;
|
||||
}
|
||||
|
||||
int main(int argc, char ** argv) {
|
||||
int main(int argc, char **argv)
|
||||
{
|
||||
gpt_params params;
|
||||
params.model = "models/llama-7B/ggml-model.bin";
|
||||
|
||||
|
|
@ -438,7 +502,8 @@ int main(int argc, char ** argv) {
|
|||
|
||||
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);
|
||||
"expect poor results\n",
|
||||
__func__, params.n_ctx);
|
||||
|
||||
if (params.seed <= 0)
|
||||
params.seed = time(NULL);
|
||||
|
|
@ -459,7 +524,8 @@ int main(int argc, char ** argv) {
|
|||
|
||||
ctx = llama_init_from_file(params.model.c_str(), lparams);
|
||||
|
||||
if (ctx == NULL) {
|
||||
if (ctx == NULL)
|
||||
{
|
||||
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
|
@ -468,8 +534,8 @@ int main(int argc, char ** argv) {
|
|||
crow::SimpleApp app;
|
||||
// app.loglevel(crow::LogLevel::Warning);
|
||||
|
||||
CROW_ROUTE(app, "/completion").methods("POST"_method)
|
||||
([¶ms, &ctx](const crow::request& req){
|
||||
CROW_ROUTE(app, "/completion").methods("POST"_method)([¶ms, &ctx](const crow::request &req)
|
||||
{
|
||||
auto body = crow::json::load(req.body);
|
||||
if (!body) return crow::response(crow::status::BAD_REQUEST);
|
||||
|
||||
|
|
@ -502,8 +568,7 @@ int main(int argc, char ** argv) {
|
|||
// Write output of LLaMA to file stream.
|
||||
run_llama(ctx, runparams, &outfile);
|
||||
|
||||
return crow::response(crow::status::OK);
|
||||
});
|
||||
return crow::response(crow::status::OK); });
|
||||
|
||||
// CROW_ROUTE(app, "/embedding").methods("POST"_method)
|
||||
// ([¶ms, &ctx](const crow::request& req){
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue