-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy pathDockerfile
More file actions
29 lines (23 loc) · 1 KB
/
Copy pathDockerfile
File metadata and controls
29 lines (23 loc) · 1 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
# FaceRecognition-Linux — Linux SDK + Docker (same repo)
# Native libs are linux/amd64; image runs on any Docker host (Win/Mac/Linux).
# Before build: download Drive folder contents into ./lib/cpu/ (see lib/README.md)
# Bookworm is glibc 2.36; native SDK libs need GLIBC_2.38+ (Debian Trixie).
FROM --platform=linux/amd64 python:3.12-slim-trixie
RUN apt-get update -y && apt-get install -y --no-install-recommends \
psmisc curl util-linux e2fsprogs libgomp1 libuuid1 zlib1g \
&& rm -rf /var/lib/apt/lists/*
WORKDIR /root/facesdk
COPY requirements.txt .
RUN pip3 install --no-cache-dir -r requirements.txt
COPY app.py sdk.py license_ux.py http_body.py run.sh ./
COPY lib ./lib/
ENV LICENSE=/root/facesdk/license.txt
ENV LD_LIBRARY_PATH=/root/facesdk/lib/cpu
# Listen on product API port 8083 (same as host publish)
ENV PORT=8083
RUN chmod +x ./run.sh \
&& test -f ./lib/cpu/libFaceRecognitionSDK.so \
&& test -f ./lib/cpu/libfar-eng.so \
&& test -f ./lib/cpu/far.fpk
CMD ["./run.sh"]
EXPOSE 8083