A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 06:45 UTC
[FIX] Fix docker build error (#1831) (#1832)
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
Allen committed
Nov 30, 2023 at 15:06 UTC
f07c1ceaa5162e2ad3b127e06b162508dc57e4db
3 files changed
+13
-1
Dockerfile
+5
@@ -18,6 +18,11 @@ RUN --mount=type=cache,target=/root/.cache/pip \
18
# image to build pytorch extensions
19
FROM dev AS build
20
21
+# install build dependencies
22
+COPY requirements-build.txt requirements-build.txt
23
+RUN --mount=type=cache,target=/root/.cache/pip \
24
+ pip install -r requirements-build.txt
25
+
26
# copy input files
27
COPY csrc csrc
28
COPY setup.py setup.py
pyproject.toml
+2
-1
@@ -1,8 +1,9 @@
1
[build-system]
2
+# Should be mirrored in requirements-build.txt
3
requires = [
4
"ninja",
5
"packaging",
5
- "setuptools",
6
+ "setuptools >= 49.4.0",
7
"torch >= 2.1.0",
8
"wheel",
9
]
requirements-build.txt
new
+6
@@ -0,0 +1,6 @@
1
+# Should be mirrored in pyproject.toml
2
+ninja
3
+packaging
4
+setuptools>=49.4.0
5
+torch>=2.1.0
6
+wheel
\ No newline at end of file