huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Preping 1.1.0

Browse Source
This commit is contained in:
Nicolas Patry 2023-09-26 20:04:38 +00:00
parent 8672cad2cb
commit 853f09035c
19 changed files with 662 additions and 766 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1255
Cargo.lock generated
View File

File diff suppressed because it is too large Load Diff

2
Cargo.toml
View File

@ -8,7 +8,7 @@ members = [
]
[workspace.package]
version = "1.0.3"
version = "1.1.0"
edition = "2021"
authors = ["Olivier Dehaene"]
homepage = "https://github.com/huggingface/text-generation-inference"

2
README.md
View File

@ -87,7 +87,7 @@ The easiest way of getting started is using the official Docker container:
model=tiiuae/falcon-7b-instruct
volume=$PWD/data # share a volume with the Docker container to avoid downloading weights every run
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:1.0.3 --model-id $model
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:1.1.0 --model-id $model
```
**Note:** To use GPUs, you need to install the [NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html). We also recommend using NVIDIA drivers with CUDA version 11.8 or higher. For running the Docker container on a machine with no GPUs or CUDA support, it is enough to remove the `--gpus all` flag and add `--disable-custom-kernels`, please note CPU is not the intended platform for this project, so performance might be subpar.

21
benchmark/Cargo.toml
View File

@ -14,18 +14,19 @@ name = "text-generation-benchmark"
path = "src/main.rs"
[dependencies]
average = "0.13"
clap = { version = "4.1.4", features = ["derive", "env"] }
crossterm = "0.26"
average = "0.14"
clap = { version = "4.4.5", features = ["derive", "env"] }
crossterm = "0.27"
float-ord = "0.3.2"
serde = {version = "1.0.142", features = ["derive"]}
serde = {version = "1.0.188", features = ["derive"]}
serde_json = "1.0"
tabled = "0.12.0"
tabled = "0.14.0"
text-generation-client = { path = "../router/client" }
thiserror = "1.0.38"
tokenizers = "0.13.3"
tokio = { version = "1.25.0", features = ["rt", "rt-multi-thread", "parking_lot", "signal", "sync"] }
tui = {package = "ratatui", version = "0.20", default-features = false, features = ["crossterm"]}
thiserror = "1.0.48"
tokenizers = { version = "0.14.0", features = ["http"] }
tokio = { version = "1.32.0", features = ["rt", "rt-multi-thread", "parking_lot", "signal", "sync", "macros"] }
tui = {package = "ratatui", version = "0.23", default-features = false, features = ["crossterm"]}
tracing = "0.1.37"
tracing-subscriber = { version = "0.3.16", features = ["json", "env-filter"] }
tracing-subscriber = { version = "0.3.17", features = ["json", "env-filter"] }
hf-hub = "0.3.1"

14
benchmark/src/app.rs
View File

@ -6,7 +6,7 @@ use tokio::sync::mpsc;
use tui::backend::Backend;
use tui::layout::{Alignment, Constraint, Direction, Layout};
use tui::style::{Color, Modifier, Style};
use tui::text::{Span, Spans};
use tui::text::{Span, Line};
use tui::widgets::{
Axis, BarChart, Block, Borders, Chart, Dataset, Gauge, GraphType, Paragraph, Tabs,
};
@ -244,7 +244,7 @@ impl App {
.batch_size
.iter()
.map(|b| {
Spans::from(vec![Span::styled(
Line::from(vec![Span::styled(
format!("Batch: {b}"),
Style::default().fg(Color::White),
)])
@ -468,7 +468,7 @@ fn latency_paragraph<'a>(latency: &mut Vec<f64>, name: &'static str) -> Paragrap
// Latency p50/p90/p99 texts
let colors = vec![Color::LightGreen, Color::LightYellow, Color::LightRed];
for (i, (name, value)) in latency_percentiles.iter().enumerate() {
let span = Spans::from(vec![Span::styled(
let span = Line::from(vec![Span::styled(
format!("{name}: {value:.2} ms"),
Style::default().fg(colors[i]),
)]);
@ -483,16 +483,16 @@ fn latency_paragraph<'a>(latency: &mut Vec<f64>, name: &'static str) -> Paragrap
}
/// Average/High/Low spans
fn statis_spans<'a>(data: &Vec<f64>, unit: &'static str) -> Vec<Spans<'a>> {
fn statis_spans<'a>(data: &Vec<f64>, unit: &'static str) -> Vec<Line<'a>> {
vec![
Spans::from(vec![Span::styled(
Line::from(vec![Span::styled(
format!(
"Average: {:.2} {unit}",
data.iter().sum::<f64>() / data.len() as f64
),
Style::default().fg(Color::LightBlue),
)]),
Spans::from(vec![Span::styled(
Line::from(vec![Span::styled(
format!(
"Lowest: {:.2} {unit}",
data.iter()
@ -501,7 +501,7 @@ fn statis_spans<'a>(data: &Vec<f64>, unit: &'static str) -> Vec<Spans<'a>> {
),
Style::default().fg(Color::Reset),
)]),
Spans::from(vec![Span::styled(
Line::from(vec![Span::styled(
format!(
"Highest: {:.2} {unit}",
data.iter()

2
benchmark/src/lib.rs
View File

@ -33,7 +33,7 @@ pub async fn run(
watermark: bool,
do_sample: bool,
client: ShardedClient,
) -> Result<(), crossterm::ErrorKind> {
) -> Result<(), std::io::Error> {
let parameters = NextTokenChooserParameters {
temperature: temperature.unwrap_or(1.0),
top_k: top_k.unwrap_or(0),

2
docs/openapi.json
View File

@ -10,7 +10,7 @@
"name": "Apache 2.0",
"url": "https://www.apache.org/licenses/LICENSE-2.0"
},
"version": "1.0.3"
"version": "1.1.0"
},
"paths": {
"/": {

2
docs/source/basic_tutorials/gated_model_access.md
View File

@ -19,6 +19,6 @@ docker run --gpus all \
--shm-size 1g \
-e HUGGING_FACE_HUB_TOKEN=$token \
-p 8080:80 \
-v $volume:/data ghcr.io/huggingface/text-generation-inference:1.0.1 \
-v $volume:/data ghcr.io/huggingface/text-generation-inference:1.1.0 \
--model-id $model
```

2
docs/source/quicktour.md
View File

@ -8,7 +8,7 @@ Let's say you want to deploy [Falcon-7B Instruct](https://huggingface.co/tiiuae/
model=tiiuae/falcon-7b-instruct
volume=$PWD/data # share a volume with the Docker container to avoid downloading weights every run
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:1.0.3 --model-id $model
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:1.1.0 --model-id $model
```
<Tip warning={true}>

2
integration-tests/pyproject.toml
View File

@ -1,6 +1,6 @@
[tool.poetry]
name = "text-generation-integration-tests"
version = "1.0.3"
version = "1.1.0"
description = "Text Generation Inference integration tests"
authors = ["Nicolas Patry <nicolas@huggingface.co>"]

16
launcher/Cargo.toml
View File

@ -7,17 +7,17 @@ authors.workspace = true
homepage.workspace = true
[dependencies]
clap = { version = "4.1.4", features = ["derive", "env"] }
ctrlc = { version = "3.2.5", features = ["termination"] }
nix = "0.26.2"
serde = { version = "1.0.152", features = ["derive"] }
serde_json = "1.0.93"
clap = { version = "4.4.5", features = ["derive", "env"] }
ctrlc = { version = "3.4.1", features = ["termination"] }
nix = "0.27.1"
serde = { version = "1.0.188", features = ["derive"] }
serde_json = "1.0.107"
tracing = "0.1.37"
tracing-subscriber = { version = "0.3.16", features = ["json", "env-filter"] }
tracing-subscriber = { version = "0.3.17", features = ["json", "env-filter"] }
[dev-dependencies]
float_eq = "1.0.1"
reqwest = { version = "0.11.14", features = ["blocking", "json"] }
reqwest = { version = "0.11.20", features = ["blocking", "json"] }
[build-dependencies]
vergen = { version = "8.0.0", features = ["build", "cargo", "git", "gitcl", "rustc", "si"] }
vergen = { version = "8.2.5", features = ["build", "cargo", "git", "gitcl", "rustc", "si"] }

46
router/Cargo.toml
View File

@ -15,35 +15,37 @@ name = "text-generation-router"
path = "src/main.rs"
[dependencies]
async-stream = "0.3.3"
axum = { version = "0.6.4", features = ["json"] }
axum-tracing-opentelemetry = "0.10.0"
async-stream = "0.3.5"
axum = { version = "0.6.20", features = ["json"] }
axum-tracing-opentelemetry = "0.14.1"
text-generation-client = { path = "client" }
clap = { version = "4.1.4", features = ["derive", "env"] }
flume = "0.10.14"
futures = "0.3.26"
metrics = "0.21.0"
clap = { version = "4.4.5", features = ["derive", "env"] }
flume = "0.11.0"
futures = "0.3.28"
metrics = "0.21.1"
metrics-exporter-prometheus = { version = "0.12.1", features = [] }
nohash-hasher = "0.2.0"
opentelemetry = { version = "0.19.0", features = ["rt-tokio"] }
opentelemetry-otlp = "0.12.0"
opentelemetry = { version = "0.20.0", features = ["rt-tokio"] }
opentelemetry-otlp = "0.13.0"
rand = "0.8.5"
reqwest = { version = "0.11.14", features = [] }
serde = "1.0.152"
serde_json = "1.0.93"
thiserror = "1.0.38"
tokenizers = "0.13.3"
tokio = { version = "1.25.0", features = ["rt", "rt-multi-thread", "parking_lot", "signal", "sync"] }
tower-http = { version = "0.4.0", features = ["cors"] }
reqwest = { version = "0.11.20", features = [] }
serde = "1.0.188"
serde_json = "1.0.107"
thiserror = "1.0.48"
tokenizers = { version = "0.14.0", features = ["http"] }
tokio = { version = "1.32.0", features = ["rt", "rt-multi-thread", "parking_lot", "signal", "sync"] }
tower-http = { version = "0.4.4", features = ["cors"] }
tracing = "0.1.37"
tracing-opentelemetry = "0.19.0"
tracing-subscriber = { version = "0.3.16", features = ["json", "env-filter"] }
utoipa = { version = "3.0.1", features = ["axum_extras"] }
utoipa-swagger-ui = { version = "3.0.2", features = ["axum"] }
ngrok = { version = "0.12.3", features = ["axum"], optional = true }
tracing-opentelemetry = "0.21.0"
tracing-subscriber = { version = "0.3.17", features = ["json", "env-filter"] }
utoipa = { version = "3.5.0", features = ["axum_extras"] }
utoipa-swagger-ui = { version = "3.1.5", features = ["axum"] }
ngrok = { version = "0.13.1", features = ["axum"], optional = true }
hf-hub = "0.3.1"
init-tracing-opentelemetry = { version = "0.14.1", features = ["opentelemetry-otlp"] }
[build-dependencies]
vergen = { version = "8.0.0", features = ["build", "git", "gitcl"] }
vergen = { version = "8.2.5", features = ["build", "git", "gitcl"] }
[features]
default = ["ngrok"]

10
router/client/Cargo.toml
View File

@ -8,13 +8,13 @@ homepage.workspace = true
[dependencies]
futures = "^0.3"
grpc-metadata = { path = "../grpc-metadata" }
prost = "^0.11"
prost = "^0.12"
thiserror = "^1.0"
tokio = { version = "^1.25", features = ["sync"] }
tonic = "^0.9"
tokio = { version = "^1.32", features = ["sync"] }
tonic = "^0.10"
tower = "^0.4"
tracing = "^0.1"
[build-dependencies]
tonic-build = "0.9.2"
prost-build = "0.11.6"
tonic-build = "0.10.1"
prost-build = "0.12.1"

6
router/grpc-metadata/Cargo.toml
View File

@ -4,7 +4,7 @@ version = "0.1.0"
edition = "2021"
[dependencies]
opentelemetry = "^0.19"
tonic = "^0.9"
opentelemetry = "^0.20"
tonic = "^0.10"
tracing = "^0.1"
tracing-opentelemetry = "^0.19"
tracing-opentelemetry = "^0.21"

2
router/src/main.rs
View File

@ -324,7 +324,7 @@ fn init_logging(otlp_endpoint: Option<String>, json_output: bool) {
if let Ok(tracer) = tracer {
layers.push(tracing_opentelemetry::layer().with_tracer(tracer).boxed());
axum_tracing_opentelemetry::init_propagator().unwrap();
init_tracing_opentelemetry::init_propagator().unwrap();
};
}

10
router/src/server.rs
View File

@ -13,7 +13,7 @@ use axum::response::sse::{Event, KeepAlive, Sse};
use axum::response::{IntoResponse, Response};
use axum::routing::{get, post};
use axum::{http, Json, Router};
use axum_tracing_opentelemetry::opentelemetry_tracing_layer;
use axum_tracing_opentelemetry::middleware::OtelAxumLayer;
use futures::stream::StreamExt;
use futures::Stream;
use metrics_exporter_prometheus::{Matcher, PrometheusBuilder, PrometheusHandle};
@ -396,7 +396,7 @@ async fn generate_stream(
// StreamResponse
let stream_token = StreamResponse {
token,
top_tokens: top_tokens,
top_tokens,
generated_text: None,
details: None,
};
@ -458,7 +458,7 @@ async fn generate_stream(
let stream_token = StreamResponse {
token,
top_tokens: top_tokens,
top_tokens,
generated_text: Some(output_text),
details
};
@ -695,7 +695,7 @@ pub async fn run(
.layer(Extension(compat_return_full_text))
.layer(Extension(infer))
.layer(Extension(prom_handle.clone()))
.layer(opentelemetry_tracing_layer())
.layer(OtelAxumLayer::default())
.layer(cors_layer);
if ngrok {
@ -792,7 +792,7 @@ async fn shutdown_signal() {
impl From<i32> for FinishReason {
fn from(finish_reason: i32) -> Self {
let finish_reason = text_generation_client::FinishReason::from_i32(finish_reason).unwrap();
let finish_reason = text_generation_client::FinishReason::try_from(finish_reason).unwrap();
match finish_reason {
text_generation_client::FinishReason::Length => FinishReason::Length,
text_generation_client::FinishReason::EosToken => FinishReason::EndOfSequenceToken,

2
router/src/validation.rs
View File

@ -276,7 +276,7 @@ impl Validation {
truncate: truncate.unwrap_or(self.max_input_length) as u32,
parameters,
stopping_parameters,
top_n_tokens: top_n_tokens,
top_n_tokens,
})
}

2
server/pyproject.toml
View File

@ -1,6 +1,6 @@
[tool.poetry]
name = "text-generation-server"
version = "1.0.3"
version = "1.1.0"
description = "Text Generation Inference Python gRPC Server"
authors = ["Olivier Dehaene <olivier@huggingface.co>"]

28
server/requirements.txt
View File

@ -9,19 +9,19 @@ certifi==2023.7.22 ; python_version >= "3.9" and python_version < "3.13"
charset-normalizer==3.2.0 ; python_version >= "3.9" and python_version < "3.13"
click==8.1.7 ; python_version >= "3.9" and python_version < "3.13"
colorama==0.4.6 ; python_version >= "3.9" and python_version < "3.13" and (sys_platform == "win32" or platform_system == "Windows")
datasets==2.14.4 ; python_version >= "3.9" and python_version < "3.13"
datasets==2.14.5 ; python_version >= "3.9" and python_version < "3.13"
deprecated==1.2.14 ; python_version >= "3.9" and python_version < "3.13"
dill==0.3.7 ; python_version >= "3.9" and python_version < "3.13"
einops==0.6.1 ; python_version >= "3.9" and python_version < "3.13"
filelock==3.12.3 ; python_version >= "3.9" and python_version < "3.13"
filelock==3.12.4 ; python_version >= "3.9" and python_version < "3.13"
frozenlist==1.4.0 ; python_version >= "3.9" and python_version < "3.13"
fsspec==2023.6.0 ; python_version >= "3.9" and python_version < "3.13"
fsspec[http]==2023.6.0 ; python_version >= "3.9" and python_version < "3.13"
googleapis-common-protos==1.60.0 ; python_version >= "3.9" and python_version < "3.13"
grpc-interceptor==0.15.3 ; python_version >= "3.9" and python_version < "3.13"
grpcio-reflection==1.57.0 ; python_version >= "3.9" and python_version < "3.13"
grpcio-status==1.57.0 ; python_version >= "3.9" and python_version < "3.13"
grpcio==1.57.0 ; python_version >= "3.9" and python_version < "3.13"
grpcio-reflection==1.58.0 ; python_version >= "3.9" and python_version < "3.13"
grpcio-status==1.58.0 ; python_version >= "3.9" and python_version < "3.13"
grpcio==1.58.0 ; python_version >= "3.9" and python_version < "3.13"
hf-transfer==0.1.3 ; python_version >= "3.9" and python_version < "3.13"
huggingface-hub==0.16.4 ; python_version >= "3.9" and python_version < "3.13"
idna==3.4 ; python_version >= "3.9" and python_version < "3.13"
@ -32,7 +32,7 @@ mpmath==1.3.0 ; python_version >= "3.9" and python_version < "3.13"
multidict==6.0.4 ; python_version >= "3.9" and python_version < "3.13"
multiprocess==0.70.15 ; python_version >= "3.9" and python_version < "3.13"
networkx==3.1 ; python_version >= "3.9" and python_version < "3.13"
numpy==1.25.2 ; python_version >= "3.9" and python_version < "3.13"
numpy==1.26.0 ; python_version >= "3.9" and python_version < "3.13"
opentelemetry-api==1.15.0 ; python_version >= "3.9" and python_version < "3.13"
opentelemetry-exporter-otlp-proto-grpc==1.15.0 ; python_version >= "3.9" and python_version < "3.13"
opentelemetry-exporter-otlp-proto-http==1.15.0 ; python_version >= "3.9" and python_version < "3.13"
@ -43,32 +43,32 @@ opentelemetry-proto==1.15.0 ; python_version >= "3.9" and python_version < "3.13
opentelemetry-sdk==1.15.0 ; python_version >= "3.9" and python_version < "3.13"
opentelemetry-semantic-conventions==0.36b0 ; python_version >= "3.9" and python_version < "3.13"
packaging==23.1 ; python_version >= "3.9" and python_version < "3.13"
pandas==2.0.3 ; python_version >= "3.9" and python_version < "3.13"
pandas==2.1.1 ; python_version >= "3.9" and python_version < "3.13"
peft==0.4.0 ; python_version >= "3.9" and python_version < "3.13"
pillow==10.0.0 ; python_version >= "3.9" and python_version < "3.13"
protobuf==4.24.2 ; python_version >= "3.9" and python_version < "3.13"
pillow==10.0.1 ; python_version >= "3.9" and python_version < "3.13"
protobuf==4.24.3 ; python_version >= "3.9" and python_version < "3.13"
psutil==5.9.5 ; python_version >= "3.9" and python_version < "3.13"
pyarrow==13.0.0 ; python_version >= "3.9" and python_version < "3.13"
python-dateutil==2.8.2 ; python_version >= "3.9" and python_version < "3.13"
pytz==2023.3 ; python_version >= "3.9" and python_version < "3.13"
pytz==2023.3.post1 ; python_version >= "3.9" and python_version < "3.13"
pyyaml==6.0.1 ; python_version >= "3.9" and python_version < "3.13"
regex==2023.8.8 ; python_version >= "3.9" and python_version < "3.13"
requests==2.31.0 ; python_version >= "3.9" and python_version < "3.13"
safetensors==0.3.3 ; python_version >= "3.9" and python_version < "3.13"
scipy==1.11.2 ; python_version >= "3.9" and python_version < "3.13"
sentencepiece==0.1.99 ; python_version >= "3.9" and python_version < "3.13"
setuptools==68.1.2 ; python_version >= "3.9" and python_version < "3.13"
setuptools==68.2.2 ; python_version >= "3.9" and python_version < "3.13"
six==1.16.0 ; python_version >= "3.9" and python_version < "3.13"
sympy==1.12 ; python_version >= "3.9" and python_version < "3.13"
texttable==1.6.7 ; python_version >= "3.9" and python_version < "3.13"
tokenizers==0.13.3 ; python_version >= "3.9" and python_version < "3.13"
torch==2.0.1 ; python_version >= "3.9" and python_version < "3.13"
tqdm==4.66.1 ; python_version >= "3.9" and python_version < "3.13"
transformers==4.32.1 ; python_version >= "3.9" and python_version < "3.13"
transformers==4.33.2 ; python_version >= "3.9" and python_version < "3.13"
typer==0.6.1 ; python_version >= "3.9" and python_version < "3.13"
typing-extensions==4.7.1 ; python_version >= "3.9" and python_version < "3.13"
typing-extensions==4.8.0 ; python_version >= "3.9" and python_version < "3.13"
tzdata==2023.3 ; python_version >= "3.9" and python_version < "3.13"
urllib3==2.0.4 ; python_version >= "3.9" and python_version < "3.13"
urllib3==2.0.5 ; python_version >= "3.9" and python_version < "3.13"
win32-setctime==1.1.0 ; python_version >= "3.9" and python_version < "3.13" and sys_platform == "win32"
wrapt==1.15.0 ; python_version >= "3.9" and python_version < "3.13"
xxhash==3.3.0 ; python_version >= "3.9" and python_version < "3.13"