llama.cpp/examples/simple/simple.cpp
slaren 16bc66d947
llama.cpp : split llama_context_params into model and context params (#3301)
* llama.cpp : split llama_context_params into model and context params

ggml-ci

* fix metal build

* fix freq_base/scale default to model value

* llama-bench : keep the same model between tests when possible

* move n_threads to llama_context_params, add n_threads_batch

* fix mpi build

* remove kv_size(), cuda scratch fixes

* remove low-vram option

* add n_threads_batch to system info, refactor to get_system_info()

* add documentation about --threads-batch to the READMEs

* llama-bench fix

* main : fix rope freq/scale warning

* llama.cpp : add llama_get_model
common : add llama_tokenize from model

* remove duplicated ctx/model functions

ggml-ci

* cuda : print total VRAM used
2023-09-28 22:42:38 +03:00

193 lines
5.2 KiB
C++

#include "common.h"
#include "llama.h"
#include <cmath>
#include <cstdio>
#include <string>
#include <vector>
int main(int argc, char ** argv) {
gpt_params params;
if (argc == 1 || argv[1][0] == '-') {
printf("usage: %s MODEL_PATH [PROMPT]\n" , argv[0]);
return 1 ;
}
if (argc >= 2) {
params.model = argv[1];
}
if (argc >= 3) {
params.prompt = argv[2];
}
if (params.prompt.empty()) {
params.prompt = "Hello my name is";
}
// total length of the sequence including the prompt
const int n_len = 32;
// init LLM
llama_backend_init(params.numa);
// initialize the model
llama_model_params model_params = llama_model_default_params();
// model_params.n_gpu_layers = 99; // offload all layers to the GPU
llama_model * model = llama_load_model_from_file(params.model.c_str(), model_params);
if (model == NULL) {
fprintf(stderr , "%s: error: unable to load model\n" , __func__);
return 1;
}
// initialize the context
llama_context_params ctx_params = llama_context_default_params();
ctx_params.seed = 1234;
ctx_params.n_ctx = 2048;
ctx_params.n_threads = params.n_threads;
ctx_params.n_threads_batch = params.n_threads_batch == -1 ? params.n_threads : params.n_threads_batch;
llama_context * ctx = llama_new_context_with_model(model, ctx_params);
if (ctx == NULL) {
fprintf(stderr , "%s: error: failed to create the llama_context\n" , __func__);
return 1;
}
// tokenize the prompt
std::vector<llama_token> tokens_list;
tokens_list = ::llama_tokenize(ctx, params.prompt, true);
const int n_ctx = llama_n_ctx(ctx);
const int n_kv_req = tokens_list.size() + (n_len - tokens_list.size());
LOG_TEE("\n%s: n_len = %d, n_ctx = %d, n_kv_req = %d\n", __func__, n_len, n_ctx, n_kv_req);
// make sure the KV cache is big enough to hold all the prompt and generated tokens
if (n_kv_req > n_ctx) {
LOG_TEE("%s: error: n_kv_req > n_ctx, the required KV cache size is not big enough\n", __func__);
LOG_TEE("%s: either reduce n_parallel or increase n_ctx\n", __func__);
return 1;
}
// print the prompt token-by-token
fprintf(stderr, "\n");
for (auto id : tokens_list) {
fprintf(stderr, "%s", llama_token_to_piece(ctx, id).c_str());
}
fflush(stderr);
// create a llama_batch with size 512
// we use this object to submit token data for decoding
llama_batch batch = llama_batch_init(512, 0);
// evaluate the initial prompt
batch.n_tokens = tokens_list.size();
for (int32_t i = 0; i < batch.n_tokens; i++) {
batch.token[i] = tokens_list[i];
batch.pos[i] = i;
batch.seq_id[i] = 0;
batch.logits[i] = false;
}
// llama_decode will output logits only for the last token of the prompt
batch.logits[batch.n_tokens - 1] = true;
if (llama_decode(ctx, batch) != 0) {
LOG_TEE("%s: llama_decode() failed\n", __func__);
return 1;
}
// main loop
int n_cur = batch.n_tokens;
int n_decode = 0;
const auto t_main_start = ggml_time_us();
while (n_cur <= n_len) {
// sample the next token
{
auto n_vocab = llama_n_vocab(model);
auto * logits = llama_get_logits_ith(ctx, batch.n_tokens - 1);
std::vector<llama_token_data> candidates;
candidates.reserve(n_vocab);
for (llama_token token_id = 0; token_id < n_vocab; token_id++) {
candidates.emplace_back(llama_token_data{ token_id, logits[token_id], 0.0f });
}
llama_token_data_array candidates_p = { candidates.data(), candidates.size(), false };
// sample the most likely token
const llama_token new_token_id = llama_sample_token_greedy(ctx, &candidates_p);
// is it an end of stream?
if (new_token_id == llama_token_eos(ctx) || n_cur == n_len) {
LOG_TEE("\n");
break;
}
LOG_TEE("%s", llama_token_to_piece(ctx, new_token_id).c_str());
fflush(stdout);
// prepare the next batch
batch.n_tokens = 0;
// push this new token for next evaluation
batch.token [batch.n_tokens] = new_token_id;
batch.pos [batch.n_tokens] = n_cur;
batch.seq_id[batch.n_tokens] = 0;
batch.logits[batch.n_tokens] = true;
batch.n_tokens += 1;
n_decode += 1;
}
n_cur += 1;
// evaluate the current batch with the transformer model
if (llama_decode(ctx, batch)) {
fprintf(stderr, "%s : failed to eval, return code %d\n", __func__, 1);
return 1;
}
}
LOG_TEE("\n");
const auto t_main_end = ggml_time_us();
LOG_TEE("%s: decoded %d tokens in %.2f s, speed: %.2f t/s\n",
__func__, n_decode, (t_main_end - t_main_start) / 1000000.0f, n_decode / ((t_main_end - t_main_start) / 1000000.0f));
llama_print_timings(ctx);
fprintf(stderr, "\n");
llama_batch_free(batch);
llama_free(ctx);
llama_free_model(model);
llama_backend_free();
return 0;
}