Skip to content

Commit

Permalink
dolly : add interactive prompt and port mode (#319)
Browse files Browse the repository at this point in the history 
* update basic function to execute prompt

* try to factor our prediciton loop

* update code

* update prompt things

* only render at the end

* add basic server port

* refactor

* fix client file descriptor

* undo common.h style changes

* undo sytle changes to main.cpp

* fix check for interactive port
  • Loading branch information
@koogle
koogle committed Jul 2, 2023
1 parent eea1d89 commit 965568d
Show file tree
Hide file tree
Showing 3 changed files with 197 additions and 64 deletions.
5 changes: 5 additions & 0 deletions examples/common.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,11 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
params.n_batch = std::stoi(argv[++i]);
} else if (arg == "-m" || arg == "--model") {
params.model = argv[++i];
} else if (arg == "-i" || arg == "--interactive") {
params.interactive = true;
} else if (arg == "-ip" || arg == "--interactive-port") {
params.interactive = true;
params.interactive_port = std::stoi(argv[++i]);
} else if (arg == "-h" || arg == "--help") {
gpt_print_usage(argc, argv, params);
exit(0);
Expand Down
3 changes: 3 additions & 0 deletions examples/common.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,9 @@ struct gpt_params {
std::string model = "models/gpt-2-117M/ggml-model.bin"; // model path
std::string prompt = "";
std::string token_test = "";

bool interactive = false;
int interactive_port = -1;
};

bool gpt_params_parse(int argc, char ** argv, gpt_params & params);
Expand Down
253 changes: 189 additions & 64 deletions examples/dolly-v2/main.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,10 +9,16 @@
#include <cstring>
#include <cinttypes>
#include <fstream>
#include <iostream>
#include <map>
#include <string>
#include <vector>

#include <arpa/inet.h>
#include <netinet/in.h>
#include <sys/socket.h>
#include <unistd.h>

#if defined(_MSC_VER)
#pragma warning(disable: 4244 4267) // possible loss of data
#endif
Expand Down Expand Up @@ -671,61 +677,25 @@ bool dollyv2_eval(
return true;
}

int main(int argc, char ** argv) {
ggml_time_init();

const int64_t t_main_start_us = ggml_time_us();

gpt_params params;
params.model = "models/dolly-v2-3b/ggml-model-f16.bin";

if (gpt_params_parse(argc, argv, params) == false) {
return 1;
}

if (params.seed < 0) {
params.seed = time(NULL);
}

printf("%s: seed = %d\n", __func__, params.seed);

std::mt19937 rng(params.seed);
if (params.prompt.empty()) {
params.prompt = gpt_random_prompt(rng);
}

const std::string prompt = prompt_for_generation(params.prompt);

int64_t t_load_us = 0;

gpt_vocab vocab;
dollyv2_model model;

// load the model
{
const int64_t t_start_us = ggml_time_us();

if (!dollyv2_model_load(params.model, model, vocab)) {
fprintf(stderr, "%s: failed to load model from '%s'\n", __func__, params.model.c_str());
return 1;
}

t_load_us = ggml_time_us() - t_start_us;

test_gpt_tokenizer(vocab, params.token_test);
}

int n_past = 0;

int64_t t_sample_us = 0;
int64_t t_predict_us = 0;

std::string execute_prompt(
const dollyv2_model &model,
gpt_vocab &vocab,
const std::string &prompt,
gpt_params &params,
std::mt19937 &rng,
int64_t t_load_us,
int64_t t_sample_us,
int64_t t_predict_us,
size_t mem_per_token,
int n_past,
bool stream_response_to_cout = false) {
std::string output = "";
std::vector<float> logits;

// tokenize the prompt
std::vector<gpt_vocab::id> embd_inp = ::gpt_tokenize(vocab, prompt);

params.n_predict = std::min(params.n_predict, model.hparams.n_ctx - (int) embd_inp.size());
params.n_predict = std::min(params.n_predict, model.hparams.n_ctx - (int)embd_inp.size());

printf("%s: number of tokens in prompt = %zu\n", __func__, embd_inp.size());
for (int i = 0; i < embd_inp.size(); i++) {
Expand All @@ -735,9 +705,7 @@ int main(int argc, char ** argv) {

std::vector<gpt_vocab::id> embd;

// determine the required inference memory per token:
size_t mem_per_token = 0;
dollyv2_eval(model, params.n_threads, 0, { 0, 1, 2, 3 }, logits, mem_per_token);
dollyv2_eval(model, params.n_threads, 0, {0, 1, 2, 3}, logits, mem_per_token);

const int32_t end_token = vocab.token_to_id["### End"];

Expand All @@ -748,7 +716,7 @@ int main(int argc, char ** argv) {

if (!dollyv2_eval(model, params.n_threads, n_past, embd, logits, mem_per_token)) {
printf("Failed to predict\n");
return 1;
return output;
}

t_predict_us += ggml_time_us() - t_start_us;
Expand All @@ -759,9 +727,9 @@ int main(int argc, char ** argv) {

if (i >= embd_inp.size()) {
// sample next token
const int top_k = params.top_k;
const int top_k = params.top_k;
const float top_p = params.top_p;
const float temp = params.temp;
const float temp = params.temp;

const int n_vocab = model.hparams.n_vocab;

Expand All @@ -777,7 +745,6 @@ int main(int argc, char ** argv) {

// add it to the context
embd.push_back(id);

} else {
// if here, it means we are still processing the input prompt
for (int k = i; k < embd_inp.size(); k++) {
Expand All @@ -791,29 +758,187 @@ int main(int argc, char ** argv) {

// display text
for (auto id : embd) {
printf("%s", vocab.id_to_token[id].c_str());
output += vocab.id_to_token[id];
if (stream_response_to_cout) {
printf("%s", vocab.id_to_token[id].c_str());
}
}
if (stream_response_to_cout) {
fflush(stdout);
}
fflush(stdout);

// end of text token
if (embd.back() == 0 || (end_token > 0 && embd.back() == end_token)) {
break;
return output;
}
}
return output;
}

int setup_port(const int port) {
int sockfd = socket(AF_INET, SOCK_STREAM, 0);
if (sockfd < 0) {
std::cerr << "Failed to create socket\n";
return -1;
}

sockaddr_in servaddr;
std::memset(&servaddr, 0, sizeof(servaddr));

servaddr.sin_family = AF_INET;
servaddr.sin_addr.s_addr = htonl(INADDR_ANY);
servaddr.sin_port = htons(port);

if (bind(sockfd, (struct sockaddr *)&servaddr, sizeof(servaddr)) < 0) {
std::cerr << "Failed to bind to port\n";
return -1;
}

if (listen(sockfd, 10) < 0) {
std::cerr << "Failed to listen on socket\n";
return -1;
}
return sockfd;
}

std::string read_from_port(int sockfd, int clientfd) {
if (clientfd < 0) {
std::cerr << "Failed to accept new connection\n";
return "";
}

char buffer[4096];
std::memset(buffer, 0, sizeof(buffer));

if (read(clientfd, buffer, sizeof(buffer)) < 0) {
std::cerr << "Failed to read from client\n";
} else {
std::cout << "Received: " << buffer;
return std::string(buffer);
}
return std::string("");
}

int main(int argc, char ** argv) {
ggml_time_init();

const int64_t t_main_start_us = ggml_time_us();

gpt_params params;
params.model = "models/dolly-v2-3b/ggml-model-f16.bin";

if (gpt_params_parse(argc, argv, params) == false) {
return 1;
}

if (params.seed < 0) {
params.seed = time(NULL);
}

printf("%s: seed = %d\n", __func__, params.seed);

std::mt19937 rng(params.seed);

int64_t t_load_us = 0;
int64_t t_sample_us = 0;
int64_t t_predict_us = 0;

// determine the required inference memory per token:
size_t mem_per_token = 0;

int n_past = 0;

gpt_vocab vocab;
dollyv2_model model;

// load the model
{
const int64_t t_start_us = ggml_time_us();

if (!dollyv2_model_load(params.model, model, vocab)) {
fprintf(stderr, "%s: failed to load model from '%s'\n", __func__, params.model.c_str());
return 1;
}

t_load_us = ggml_time_us() - t_start_us;

test_gpt_tokenizer(vocab, params.token_test);
}

int sockfd;
if (params.interactive_port != -1) {
sockfd = setup_port(params.interactive_port);
if (sockfd == -1) {
return 1;
}
fprintf(stdout, "Model is ready on port %i\n", params.interactive_port);
fflush(stdout);
}

if (params.interactive or params.interactive_port != -1) {
while (true) {
std::string prompt_input;
int clientfd;
if (params.interactive_port != -1) {
sockaddr_in clientaddr;
socklen_t clientaddrlen = sizeof(clientaddr);
clientfd = accept(sockfd, (struct sockaddr *)&clientaddr, &clientaddrlen);
prompt_input = read_from_port(sockfd, clientfd);
} else {
printf("Please enter your quesiton:\n>");
fflush(stdout);

std::getline(std::cin, prompt_input);
}

if (strcmp(prompt_input.c_str(), "exit") == 0) {
break;
}

const std::string prompt = prompt_for_generation(prompt_input);
// call the model
const std::string response = execute_prompt(model, vocab, prompt, params, rng, t_load_us, t_sample_us, t_predict_us, mem_per_token, n_past, true);

if (params.interactive_port != -1) {
if (write(clientfd, response.c_str(), response.size()) < 0) {
std::cerr << "Failed to write to client\n";
}

if (close(clientfd) < 0) {
std::cerr << "Failed to close client socket\n";
}
}
else {
printf("%s\n\n", response.c_str());
}
fflush(stdout);
}
} else {
if (params.prompt.empty()) {
params.prompt = gpt_random_prompt(rng);
}

const std::string prompt = prompt_for_generation(params.prompt);
execute_prompt(model, vocab, prompt, params, rng, t_load_us, t_sample_us, t_predict_us, mem_per_token, n_past, true);
}

// report timing
{
const int64_t t_main_end_us = ggml_time_us();

printf("\n\n");
printf("%s: mem per token = %8zu bytes\n", __func__, mem_per_token);
printf("%s: load time = %8.2f ms\n", __func__, t_load_us/1000.0f);
printf("%s: sample time = %8.2f ms\n", __func__, t_sample_us/1000.0f);
printf("%s: predict time = %8.2f ms / %.2f ms per token\n", __func__, t_predict_us/1000.0f, t_predict_us/1000.0f/n_past);
printf("%s: total time = %8.2f ms\n", __func__, (t_main_end_us - t_main_start_us)/1000.0f);
printf("%s: load time = %8.2f ms\n", __func__, t_load_us / 1000.0f);
printf("%s: sample time = %8.2f ms\n", __func__, t_sample_us / 1000.0f);
printf("%s: predict time = %8.2f ms / %.2f ms per token\n", __func__, t_predict_us / 1000.0f, t_predict_us / 1000.0f / n_past);
printf("%s: total time = %8.2f ms\n", __func__, (t_main_end_us - t_main_start_us) / 1000.0f);
}

ggml_free(model.ctx);

if (params.interactive_port != -1 && close(sockfd) < 0) {
std::cerr << "Failed to close server socket\n";
}

return 0;
}

0 comments on commit 965568d

Please sign in to comment.