[Bug] cogvlm2支持的问题

[tdf1995]

Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.

Describe the bug

2024-09-06 02:46:34,547 - lmdeploy - WARNING - Fallback to pytorch engine because /root/.cache/huggingface/hub/models--THUDM--cogvlm2-llama3-chinese-chat-19B/snapshots/9ab2ccb0e7b7db52e6ff60c50f12082af3ce12ad not supported by turbomind engine.
2024-09-06 02:46:37,662 - lmdeploy - ERROR - matching vision model: CogVLMVisionModel failed
Traceback (most recent call last):
File "/usr/bin/lmdeploy", line 8, in
sys.exit(run())
File "/usr/lib/python3.10/site-packages/lmdeploy/cli/entrypoint.py", line 36, in run
args.run(args)
File "/usr/lib/python3.10/site-packages/lmdeploy/cli/serve.py", line 298, in api_server
run_api_server(args.model_path,
File "/usr/lib/python3.10/site-packages/lmdeploy/serve/openai/api_server.py", line 1285, in serve
VariableInterface.async_engine = pipeline_class(
File "/usr/lib/python3.10/site-packages/lmdeploy/serve/vl_async_engine.py", line 21, in init
self.vl_encoder = ImageEncoder(model_path,
File "/usr/lib/python3.10/site-packages/lmdeploy/vl/engine.py", line 85, in init
self.model = load_vl_model(model_path, backend_config=backend_config)
File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/builder.py", line 55, in load_vl_model
return module(**kwargs)
File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/base.py", line 31, in init
self.build_model()
File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/cogvlm.py", line 48, in build_model
device_map = infer_auto_device_map(
File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 1238, in infer_auto_device_map
tied_parameters = find_tied_parameters(model)
File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 708, in find_tied_parameters
all_named_parameters = {name: param for name, param in _get_named_parameters(model, remove_duplicate=False)}
File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 708, in
all_named_parameters = {name: param for name, param in _get_named_parameters(model, remove_duplicate=False)}
File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 667, in _get_named_parameters
members = module._parameters.items()
AttributeError: 'NoneType' object has no attribute '_parameters'

Reproduction

lmdeploy serve api_server THUDM/cogvlm2-llama3-chinese-chat-19B --server-port 23333

Environment

sys.platform: linux
Python: 3.10.5 (main, Mar 28 2024, 16:07:02) [GCC 7.5.0]
CUDA available: True
MUSA available: False
numpy_random_seed: 2147483648
GPU 0,1,2,3: Tesla T4
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 11.8, V11.8.89
GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
PyTorch: 2.3.0+cu118
PyTorch compiling details: PyTorch built with:
  - GCC 9.3
  - C++ Version: 201703
  - Intel(R) oneAPI Math Kernel Library Version 2022.2-Product Build 20220804 for Intel(R) 64 architect                                                                                                                                                                ure applications
  - Intel(R) MKL-DNN v3.3.6 (Git Hash 86e6af5974177e513fd3fee58425e1063e7f1361)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: AVX512
  - CUDA Runtime 11.8
  - NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-g                                                                                                                                                                encode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_                                                                                                                                                                80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_90,cod                                                                                                                                                                e=sm_90
  - CuDNN 8.7
  - Magma 2.6.1
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.8, CUDNN_VERSION=8.7.0, CXX_COMP                                                                                                                                                                ILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvi                                                                                                                                                                sibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -D                                                                                                                                                                USE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wext                                                                                                                                                                ra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-i                                                                                                                                                                nitializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-f                                                                                                                                                                unction -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-o                                                                                                                                                                verride -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-missing-braces -fdiagnostics-colo                                                                                                                                                                r=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapp                                                                                                                                                                ing-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PER                                                                                                                                                                F_WITH_AVX512=1, TORCH_VERSION=2.3.0, USE_CUDA=ON, USE_CUDNN=ON, USE_CUSPARSELT=1, USE_EXCEPTION_PTR=1,                                                                                                                                                                 USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNP                                                                                                                                                                ACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF,

TorchVision: 0.18.0+cu118
LMDeploy: 0.5.3+
transformers: 4.41.2
gradio: 4.25.0
fastapi: 0.110.1
pydantic: 2.6.4
triton: 2.3.0
NVIDIA Topology:
        GPU0    GPU1    GPU2    GPU3    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      SYS     SYS     SYS     0-19,40-59      0               N/A
GPU1    SYS      X      SYS     SYS     0-19,40-59      0               N/A
GPU2    SYS     SYS      X      SYS     20-39,60-79     1               N/A
GPU3    SYS     SYS     SYS      X      20-39,60-79     1               N/A

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA                                                                                                                                                                 node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

Error traceback

2024-09-06 02:46:34,547 - lmdeploy - WARNING - Fallback to pytorch engine because `/root/.cache/huggingface/hub/models--THUDM--cogvlm2-llama3-chinese-chat-19B/snapshots/9ab2ccb0e7b7db52e6ff60c50f12082af3ce12ad` not supported by turbomind engine.
2024-09-06 02:46:37,662 - lmdeploy - ERROR - matching vision model: CogVLMVisionModel failed
Traceback (most recent call last):
  File "/usr/bin/lmdeploy", line 8, in <module>
    sys.exit(run())
  File "/usr/lib/python3.10/site-packages/lmdeploy/cli/entrypoint.py", line 36, in run
    args.run(args)
  File "/usr/lib/python3.10/site-packages/lmdeploy/cli/serve.py", line 298, in api_server
    run_api_server(args.model_path,
  File "/usr/lib/python3.10/site-packages/lmdeploy/serve/openai/api_server.py", line 1285, in serve
    VariableInterface.async_engine = pipeline_class(
  File "/usr/lib/python3.10/site-packages/lmdeploy/serve/vl_async_engine.py", line 21, in __init__
    self.vl_encoder = ImageEncoder(model_path,
  File "/usr/lib/python3.10/site-packages/lmdeploy/vl/engine.py", line 85, in __init__
    self.model = load_vl_model(model_path, backend_config=backend_config)
  File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/builder.py", line 55, in load_vl_model
    return module(**kwargs)
  File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/base.py", line 31, in __init__
    self.build_model()
  File "/usr/lib/python3.10/site-packages/lmdeploy/vl/model/cogvlm.py", line 48, in build_model
    device_map = infer_auto_device_map(
  File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 1238, in infer_auto_device_map
    tied_parameters = find_tied_parameters(model)
  File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 708, in find_tied_parameters
    all_named_parameters = {name: param for name, param in _get_named_parameters(model, remove_duplicate=False)}
  File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 708, in <dictcomp>
    all_named_parameters = {name: param for name, param in _get_named_parameters(model, remove_duplicate=False)}
  File "/usr/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 667, in _get_named_parameters
    members = module._parameters.items()
AttributeError: 'NoneType' object has no attribute '_parameters'

[RunningLeon]

refer to #2055 (comment)