Mirrors
/
hf_text-generation-inference
mirror of https://github.com/huggingface/text-generation-inference.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

feat(backend): use the new batch api from llama

This commit is contained in:
Morgan Funtowicz 2024-11-28 14:52:48 +01:00
parent 274cfce435
commit 8e89793514
2 changed files with 61 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

89
backends/llamacpp/csrc/backend.cpp
View File

@ -38,6 +38,31 @@ namespace huggingface::tgi::backends::llamacpp {
return {pSampler, llama_sampler_deleter};
}
std::expected<llama_batch, backend_error_t> get_batch_from_prompt(std::span<llama_token> prompt) {
auto batch = llama_batch_init(static_cast<int32_t>(prompt.size()), 0, 1);
std::for_each(prompt.begin(), prompt.end(), [&batch](const llama_token token) {
const auto n_token = batch.n_tokens;
batch.token[n_token] = token;
batch.pos[n_token] = n_token;
batch.n_seq_id[n_token] = 1;
batch.seq_id[n_token][0] = 1;
batch.logits[n_token] = false;
batch.n_tokens++;
});
batch.logits[batch.n_tokens - 1] = true;
return batch;
}
void update_batch_for_decoding(llama_batch &batch, llama_token token, size_t position) {
batch.n_tokens = 1;
batch.logits[0] = true;
batch.token[0] = token;
batch.pos[0] = static_cast<int32_t>(position);
}
worker_t::worker_t(std::shared_ptr<llama_model> model, const llama_context_params &&params)
: model_(model), context_(llama_new_context_with_model(model_.get(), params)) {
@ -59,44 +84,50 @@ namespace huggingface::tgi::backends::llamacpp {
auto sampler = generation_context.sampling_params.into_llama_sampler(model_.get());
// Set up the prompt
auto copy = std::vector(generation_context.input_tokens.begin(), generation_context.input_tokens.end());
auto batch = llama_batch_get_one(copy.data(), copy.size());
// Decode
auto n_decoded_tokens = 0;
for (bool generating = true; generating; ++n_decoded_tokens) {
if (auto maybe_batch = get_batch_from_prompt(generation_context.input_tokens); maybe_batch.has_value()) {
// Decode
auto batch = *maybe_batch;
auto n_decoded_tokens = 0;
const auto prompt_size = generation_context.input_tokens.size();
for (bool generating = true; generating; ++n_decoded_tokens) {
#ifdef TGI_LLAMACPP_BACKEND_DEBUG
const auto start = std::chrono::steady_clock::now();
const auto status = llama_decode(context_.get(), batch);
const auto end = std::chrono::steady_clock::now();
const auto latency = std::chrono::duration_cast<std::chrono::milliseconds>(end - start);
SPDLOG_DEBUG(FMT_STRING("Successfully decoded {:d} token(s) in {}"), batch.n_tokens, latency);
const auto start = std::chrono::steady_clock::now();
const auto status = llama_decode(context_.get(), batch);
const auto end = std::chrono::steady_clock::now();
const auto latency = std::chrono::duration_cast<std::chrono::milliseconds>(end - start);
SPDLOG_DEBUG(FMT_STRING("Successfully decoded {:d} token(s) in {}"), batch.n_tokens, latency);
#else
const auto status = llama_decode(context_.get(), batch);
const auto status = llama_decode(context_.get(), batch);
#endif
batch.n_tokens = 0;
if (LLAMA_SUCCESS(status)) [[likely]] {
// Sample the new token
auto new_token_id = llama_sampler_sample(sampler.get(), context_.get(), -1);
auto is_eog = llama_token_is_eog(model_.get(), new_token_id);
auto new_token_logits = 0.0f; // TODO: return logit
batch.n_tokens = 0;
if (LLAMA_SUCCESS(status)) [[likely]] {
// Sample the new token
auto new_token_id = llama_sampler_sample(sampler.get(), context_.get(), -1);
const auto is_eog = llama_token_is_eog(model_.get(), new_token_id);
const auto new_token_logits = llama_get_logits_ith(context_.get(), -1); // TODO: return logit
// Handle termination cases
const auto has_reach_max_tokens = n_decoded_tokens >= max_new_tokens - 1;
const auto has_reach_eog = !generation_context.generation_params.ignore_eos_token & is_eog;
// Handle termination cases
const bool has_reach_max_tokens = n_decoded_tokens >= max_new_tokens - 1;
const bool has_reach_eog = !generation_context.generation_params.ignore_eos_token & is_eog;
const bool is_final = has_reach_max_tokens | has_reach_eog;
generating = !(has_reach_max_tokens | has_reach_eog);
// Bubble up the generated token if a callback is provided
const auto should_stop = callback_(new_token_id, *new_token_logits, is_final, n_decoded_tokens + 1);
// Bubble up the generated token if a callback is provided
const auto should_stop =
std::invoke(callback_, new_token_id, new_token_logits, !generating, n_decoded_tokens + 1);
generating ^= should_stop;
// Compute the continuation flag
generating = !(should_stop | is_final);
batch = llama_batch_get_one(&new_token_id, 1);
// Update the batch for the next generation
update_batch_for_decoding(batch, new_token_id, prompt_size + n_decoded_tokens);
}
}
}
return n_decoded_tokens;
llama_batch_free(batch);
return n_decoded_tokens;
} else {
return maybe_batch.error();
}
}
}

2
backends/llamacpp/csrc/backend.hpp
View File

@ -75,7 +75,7 @@ namespace huggingface::tgi::backends::llamacpp {
struct generation_context_t {
generation_params_t generation_params;
sampling_params_t sampling_params;
std::span<const llama_token> input_tokens;
std::span<llama_token> input_tokens;
};
/**