mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-24 00:12:08 +00:00
# What does this PR do? Upgrade all relevant versions and dependencies. <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this with a description of the change and which issue is fixed (if applicable). Please also include relevant motivation and context. List any dependencies (if any) that are required for this change. Once you're done, someone will review your PR shortly (see the section "Who can review?" below to tag some potential reviewers). They may suggest changes to make the code even better. If no one reviewed your PR after a week has passed, don't hesitate to post a new comment @-mentioning the same persons---sometimes notifications get lost. --> <!-- Remove if not applicable --> Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. <!-- Your PR will be replied to more quickly if you can figure out the right person to tag with @ @OlivierDehaene OR @Narsil -->
156 lines
4.1 KiB
Rust
156 lines
4.1 KiB
Rust
mod app;
|
|
mod event;
|
|
mod generation;
|
|
mod table;
|
|
mod utils;
|
|
|
|
use crate::app::App;
|
|
use crate::event::Event;
|
|
use crossterm::ExecutableCommand;
|
|
use std::io;
|
|
use text_generation_client::{NextTokenChooserParameters, ShardedClient};
|
|
use tokenizers::Tokenizer;
|
|
use tokio::sync::{broadcast, mpsc};
|
|
use tui::backend::CrosstermBackend;
|
|
use tui::Terminal;
|
|
|
|
/// Run benchmarking app
|
|
#[allow(clippy::too_many_arguments)]
|
|
pub async fn run(
|
|
tokenizer_name: String,
|
|
tokenizer: Tokenizer,
|
|
batch_size: Vec<u32>,
|
|
sequence_length: u32,
|
|
decode_length: u32,
|
|
top_n_tokens: Option<u32>,
|
|
n_runs: usize,
|
|
warmups: usize,
|
|
temperature: Option<f32>,
|
|
top_k: Option<u32>,
|
|
top_p: Option<f32>,
|
|
typical_p: Option<f32>,
|
|
repetition_penalty: Option<f32>,
|
|
watermark: bool,
|
|
do_sample: bool,
|
|
client: ShardedClient,
|
|
) -> Result<(), std::io::Error> {
|
|
let parameters = NextTokenChooserParameters {
|
|
temperature: temperature.unwrap_or(1.0),
|
|
top_k: top_k.unwrap_or(0),
|
|
top_p: top_p.unwrap_or(1.0),
|
|
typical_p: typical_p.unwrap_or(1.0),
|
|
do_sample,
|
|
seed: 0,
|
|
repetition_penalty: repetition_penalty.unwrap_or(1.0),
|
|
watermark,
|
|
};
|
|
|
|
// Initialize terminal properties
|
|
crossterm::terminal::enable_raw_mode()?;
|
|
io::stdout().execute(crossterm::terminal::EnterAlternateScreen)?;
|
|
io::stdout().execute(crossterm::cursor::Hide)?;
|
|
|
|
// Initialize terminal
|
|
let mut terminal = {
|
|
let backend = CrosstermBackend::new(io::stdout());
|
|
Terminal::new(backend)?
|
|
};
|
|
|
|
// Create message channel between generation_task and app
|
|
let (run_sender, run_receiver) = mpsc::channel(8);
|
|
// Crossterm event channel
|
|
let (event_sender, mut event_receiver) = mpsc::channel(8);
|
|
// Shutdown channel to terminate tasks
|
|
let (shutdown_sender, _) = broadcast::channel(1);
|
|
// Channel to check if tasks terminated
|
|
let (shutdown_guard_sender, mut shutdown_guard_receiver) = mpsc::channel(1);
|
|
|
|
// Create generation task
|
|
tokio::spawn(generation::generation_task(
|
|
tokenizer,
|
|
batch_size.clone(),
|
|
sequence_length,
|
|
decode_length,
|
|
top_n_tokens,
|
|
n_runs,
|
|
warmups,
|
|
parameters,
|
|
client,
|
|
run_sender,
|
|
shutdown_sender.subscribe(),
|
|
shutdown_guard_sender.clone(),
|
|
));
|
|
|
|
// Create event task
|
|
tokio::spawn(event::terminal_event_task(
|
|
250,
|
|
event_sender,
|
|
shutdown_sender.subscribe(),
|
|
shutdown_guard_sender.clone(),
|
|
));
|
|
|
|
// Drop our end of shutdown sender
|
|
drop(shutdown_guard_sender);
|
|
|
|
// Create App
|
|
let mut app = App::new(
|
|
run_receiver,
|
|
tokenizer_name.clone(),
|
|
sequence_length,
|
|
decode_length,
|
|
n_runs,
|
|
batch_size,
|
|
);
|
|
|
|
while app.running {
|
|
// Draw frame
|
|
terminal.draw(|frame| app.render(frame))?;
|
|
|
|
// Await a new event from event handling task
|
|
match event_receiver.recv().await {
|
|
None => break,
|
|
// Update app state
|
|
Some(event) => match event {
|
|
Event::Tick => app.tick(),
|
|
Event::Key(key_event) => app.handle_key_event(key_event),
|
|
_ => {}
|
|
},
|
|
}
|
|
}
|
|
|
|
// Ask tasks to shutdown
|
|
let _ = shutdown_sender.send(());
|
|
// Wait for tasks to shutdown
|
|
let _ = shutdown_guard_receiver.recv().await;
|
|
|
|
// Revert terminal to original view
|
|
io::stdout().execute(crossterm::terminal::LeaveAlternateScreen)?;
|
|
crossterm::terminal::disable_raw_mode()?;
|
|
io::stdout().execute(crossterm::cursor::Show)?;
|
|
|
|
let parameters_table = table::parameters_table(
|
|
tokenizer_name,
|
|
sequence_length,
|
|
decode_length,
|
|
top_n_tokens,
|
|
n_runs,
|
|
warmups,
|
|
temperature,
|
|
top_k,
|
|
top_p,
|
|
typical_p,
|
|
repetition_penalty,
|
|
watermark,
|
|
do_sample,
|
|
);
|
|
println!("\n{parameters_table}\n");
|
|
|
|
let latency_table = table::latency_table(&app.data);
|
|
println!("\n{latency_table}\n");
|
|
|
|
let throughput_table = table::throughput_table(&app.data);
|
|
println!("\n{throughput_table}\n");
|
|
|
|
Ok(())
|
|
}
|