huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-12 04:44:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Updating the dockerfile to make libpython discoverable at runtime too.

Browse Source
This commit is contained in:
Nicolas Patry 2024-09-11 16:45:22 +02:00
parent f518f798b7
commit 538994ced4
No known key found for this signature in database
GPG Key ID: 64AF4752B2967863
6 changed files with 28 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
Dockerfile
View File

@ -220,33 +220,33 @@ RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-ins
COPY --from=pytorch-install /opt/conda /opt/conda COPY --from=pytorch-install /opt/conda /opt/conda
# Copy build artifacts from flash attention builder # Copy build artifacts from flash attention builder
COPY --from=flash-att-builder /usr/src/flash-attention/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=flash-att-builder /usr/src/flash-attention/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
COPY --from=flash-att-builder /usr/src/flash-attention/csrc/layer_norm/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=flash-att-builder /usr/src/flash-attention/csrc/layer_norm/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
COPY --from=flash-att-builder /usr/src/flash-attention/csrc/rotary/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=flash-att-builder /usr/src/flash-attention/csrc/rotary/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from flash attention v2 builder # Copy build artifacts from flash attention v2 builder
COPY --from=flash-att-v2-builder /opt/conda/lib/python3.10/site-packages/flash_attn_2_cuda.cpython-310-x86_64-linux-gnu.so /opt/conda/lib/python3.10/site-packages COPY --from=flash-att-v2-builder /opt/conda/lib/python3.11/site-packages/flash_attn_2_cuda.cpython-311-x86_64-linux-gnu.so /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from custom kernels builder # Copy build artifacts from custom kernels builder
COPY --from=custom-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=custom-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from exllama kernels builder # Copy build artifacts from exllama kernels builder
COPY --from=exllama-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=exllama-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from exllamav2 kernels builder # Copy build artifacts from exllamav2 kernels builder
COPY --from=exllamav2-kernels-builder /usr/src/exllamav2/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=exllamav2-kernels-builder /usr/src/exllamav2/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from awq kernels builder # Copy build artifacts from awq kernels builder
COPY --from=awq-kernels-builder /usr/src/llm-awq/awq/kernels/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=awq-kernels-builder /usr/src/llm-awq/awq/kernels/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from eetq kernels builder # Copy build artifacts from eetq kernels builder
COPY --from=eetq-kernels-builder /usr/src/eetq/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=eetq-kernels-builder /usr/src/eetq/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from lorax punica kernels builder # Copy build artifacts from lorax punica kernels builder
COPY --from=lorax-punica-builder /usr/src/lorax-punica/server/punica_kernels/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=lorax-punica-builder /usr/src/lorax-punica/server/punica_kernels/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from fbgemm builder # Copy build artifacts from fbgemm builder
COPY --from=fbgemm-builder /usr/src/fbgemm/fbgemm_gpu/_skbuild/linux-x86_64-3.10/cmake-install /opt/conda/lib/python3.10/site-packages COPY --from=fbgemm-builder /usr/src/fbgemm/fbgemm_gpu/_skbuild/linux-x86_64-3.11/cmake-install /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from vllm builder # Copy build artifacts from vllm builder
COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from mamba builder # Copy build artifacts from mamba builder
COPY --from=mamba-builder /usr/src/mamba/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages COPY --from=mamba-builder /usr/src/mamba/build/lib.linux-x86_64-cpython-311/ /opt/conda/lib/python3.11/site-packages
COPY --from=mamba-builder /usr/src/causal-conv1d/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages COPY --from=mamba-builder /usr/src/causal-conv1d/build/lib.linux-x86_64-cpython-311/ /opt/conda/lib/python3.11/site-packages
COPY --from=flashinfer-builder /opt/conda/lib/python3.10/site-packages/flashinfer/ /opt/conda/lib/python3.10/site-packages/flashinfer/ COPY --from=flashinfer-builder /opt/conda/lib/python3.11/site-packages/flashinfer/ /opt/conda/lib/python3.11/site-packages/flashinfer/
# Install flash-attention dependencies # Install flash-attention dependencies
RUN pip install einops --no-cache-dir RUN pip install einops --no-cache-dir
@ -261,7 +261,9 @@ RUN cd server && \
pip install ".[bnb, accelerate, marlin, quantize, peft, outlines]" --no-cache-dir && \ pip install ".[bnb, accelerate, marlin, quantize, peft, outlines]" --no-cache-dir && \
pip install nvidia-nccl-cu12==2.22.3 pip install nvidia-nccl-cu12==2.22.3
ENV LD_PRELOAD=/opt/conda/lib/python3.10/site-packages/nvidia/nccl/lib/libnccl.so.2 ENV LD_PRELOAD=/opt/conda/lib/python3.11/site-packages/nvidia/nccl/lib/libnccl.so.2
# Required to find libpython within the rust binaries
ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/opt/conda/lib/"
# This is needed because exl2 tries to load flash-attn # This is needed because exl2 tries to load flash-attn
# And fails with our builds. # And fails with our builds.
ENV EXLLAMA_NO_FLASH_ATTN=1 ENV EXLLAMA_NO_FLASH_ATTN=1

11
Dockerfile_amd
View File

@ -174,19 +174,19 @@ ENV HF_HOME=/data \
PORT=80 PORT=80
# Copy builds artifacts from vllm builder # Copy builds artifacts from vllm builder
COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from flash attention v2 builder # Copy build artifacts from flash attention v2 builder
COPY --from=flash-att-v2-builder /usr/src/flash-attention-v2/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=flash-att-v2-builder /usr/src/flash-attention-v2/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from custom kernels builder # Copy build artifacts from custom kernels builder
COPY --from=custom-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=custom-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from exllama kernels builder # Copy build artifacts from exllama kernels builder
COPY --from=exllama-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=exllama-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Copy build artifacts from exllamav2 kernels builder # Copy build artifacts from exllamav2 kernels builder
COPY --from=exllamav2-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages COPY --from=exllamav2-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-311 /opt/conda/lib/python3.11/site-packages
# Install server # Install server
COPY proto proto COPY proto proto
@ -203,6 +203,7 @@ COPY --from=builder /usr/src/target/release-opt/text-generation-benchmark /usr/l
COPY --from=builder /usr/src/target/release-opt/text-generation-router /usr/local/bin/text-generation-router COPY --from=builder /usr/src/target/release-opt/text-generation-router /usr/local/bin/text-generation-router
# Install launcher # Install launcher
COPY --from=builder /usr/src/target/release-opt/text-generation-launcher /usr/local/bin/text-generation-launcher COPY --from=builder /usr/src/target/release-opt/text-generation-launcher /usr/local/bin/text-generation-launcher
ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/opt/conda/lib/"
# AWS Sagemaker compatible image # AWS Sagemaker compatible image
FROM base AS sagemaker FROM base AS sagemaker

1
Dockerfile_intel
View File

@ -155,6 +155,7 @@ ENV CCL_ROOT=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch
ENV I_MPI_ROOT=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch ENV I_MPI_ROOT=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch
ENV FI_PROVIDER_PATH=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib/prov:/usr/lib64/libfabric ENV FI_PROVIDER_PATH=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib/prov:/usr/lib64/libfabric
ENV LD_LIBRARY_PATH=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib:/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/lib ENV LD_LIBRARY_PATH=/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib:/opt/conda/lib/python3.10/site-packages/oneccl_bindings_for_pytorch/lib
ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/opt/conda/lib/"
# Install server # Install server
COPY proto proto COPY proto proto

2
server/tests/models/test_bloom.py
View File

@ -267,7 +267,7 @@ def test_batch_concatenate(
assert next_batch.max_input_length == 3 assert next_batch.max_input_length == 3
assert next_batch.requests[0] == next_batch_0.requests[0] assert next_batch.requests[0] == next_batch_0.requests[0]
assert next_batch.requests[1:] == next_batch_1.requests assert next_batch.requests[1:] == list(next_batch_1.requests)
assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0] assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0]
assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers

2
server/tests/models/test_causal_lm.py
View File

@ -262,7 +262,7 @@ def test_batch_concatenate(
assert next_batch.max_input_length == 3 assert next_batch.max_input_length == 3
assert next_batch.requests[0] == next_batch_0.requests[0] assert next_batch.requests[0] == next_batch_0.requests[0]
assert next_batch.requests[1:] == next_batch_1.requests assert next_batch.requests[1:] == list(next_batch_1.requests)
assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0] assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0]
assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers

2
server/tests/models/test_seq2seq_lm.py
View File

@ -281,7 +281,7 @@ def test_batch_concatenate(
assert next_batch.max_decoder_input_length == 3 assert next_batch.max_decoder_input_length == 3
assert next_batch.requests[0] == next_batch_0.requests[0] assert next_batch.requests[0] == next_batch_0.requests[0]
assert next_batch.requests[1:] == next_batch_1.requests assert next_batch.requests[1:] == list(next_batch_1.requests)
assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0] assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0]
assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers