[Bug] 如何使用python代码加载turbomind模型并对话（菜鸡请求大佬支援）

[quanfeifan]

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

我参考了https://github.com/InternLM/lmdeploy/issues/1835#issue-2369484615。
看了EngineOutput结构体，但是不太会改代码，想请大佬帮我纠正一下，如何才能正确输入与输出。我所用的模型是经过4bit量化的internlm2-7b

import json
from lmdeploy import turbomind as tm
tm_model = tm.TurboMind.from_pretrained('/root/autodl-tmp/internlm2-4b')
generator = tm_model.create_instance()

import tool
def chat(prompt):
    input_ids = tm_model.tokenizer.encode(prompt)
    for outputs in generator.stream_infer(session_id=0, input_ids=[input_ids]):
        res = outputs.token_ids
        
    response = tm_model.tokenizer.decode(res)
    print(response)
    try:
        response_dict = json.loads(response)
    except ValueError:
        response = "{\"thought\":\"" + response 
        response_dict = json.loads(response)
    prompt += response + "<|im_end|>\n"
    return prompt, response, response_dict


# system_prompt = tool.system_prompt
system_prompt  = "hello"
system_prompt_template = """<|im_start|>你是书生。<|im_end|>
<|im_start|>user
{}<|im_end|>
<|im_start|>assistant
"""

prompt, response, response_dict = chat(system_prompt_template.format(system_prompt))
print(response)

另外，我的输入是比较简单的”hello“，他的输出却要经过好几秒才给出，step=520才有结果，这正常吗？

Reproduction

import json
from lmdeploy import turbomind as tm
tm_model = tm.TurboMind.from_pretrained('/root/autodl-tmp/internlm2-4b')
generator = tm_model.create_instance()

import tool
def chat(prompt):
input_ids = tm_model.tokenizer.encode(prompt)
for outputs in generator.stream_infer(session_id=0, input_ids=[input_ids]):
res = outputs.token_ids

response = tm_model.tokenizer.decode(res)
print(response)
try:
    response_dict = json.loads(response)
except ValueError:
    response = "{\"thought\":\"" + response 
    response_dict = json.loads(response)
prompt += response + "<|im_end|>\n"
return prompt, response, response_dict

system_prompt = tool.system_prompt

system_prompt = "hello"
system_prompt_template = """<|im_start|>你是书生。<|im_end|>
<|im_start|>user
{}<|im_end|>
<|im_start|>assistant
"""

prompt, response, response_dict = chat(system_prompt_template.format(system_prompt))
print(response)

Environment

sys.platform: linux
Python: 3.10.8 (main, Nov 24 2022, 14:13:03) [GCC 11.2.0]
CUDA available: True
MUSA available: False
numpy_random_seed: 2147483648
GPU 0: NVIDIA GeForce RTX 2080 Ti
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 12.1, V12.1.105
GCC: gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
PyTorch: 2.1.2+cu121
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 architecture applications
  - Intel(R) MKL-DNN v3.1.1 (Git Hash 64f6bcbcbab628e96f33a62c3e975f8535a7bde4)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: AVX512
  - CUDA Runtime 12.1
  - NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;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_90,code=sm_90
  - CuDNN 8.9.2
  - Magma 2.6.1
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=12.1, CUDNN_VERSION=8.9.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-invalid-partial-specialization -Wno-unused-private-field -Wno-aligned-allocation-unavailable -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_DISABLE_GPU_ASSERTS=ON, TORCH_VERSION=2.1.2, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, 

TorchVision: 0.16.2+cu121
LMDeploy: 0.5.1+
transformers: 4.37.1
gradio: 3.50.2
fastapi: 0.111.1
pydantic: 2.8.2
triton: 2.1.0
NVIDIA Topology: 
        GPU0    NIC0    NIC1    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      SYS     SYS     32-63,96-127    1               N/A
NIC0    SYS      X      PIX
NIC1    SYS     PIX      X 

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

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1

Error traceback

No response

[lvhan028]

建议使用 pipeline 接口，而不是 turbomind 的接口。
pipeline 处理了对话模板，但是turbomind没有。

[quanfeifan]

建议使用 pipeline 接口，而不是 turbomind 的接口。
pipeline 处理了对话模板，但是turbomind没有。

pipeline可以实现多轮对话吗，我好像没有看到

[lvhan028]

Yes. Please read the "An example for OpenAI format prompt input:" example in the LLM pipeline user guide

[quanfeifan]

您好，我其实想用python代码达到lmdeploy chat ./model_path的效果，能够在命令行持续输入，此外我还想提取parse大模型的输出（这一点lmdeploy chat .....似乎做不到），请问您有什么方法能做到这一点吗

[zhyncs]

@quanfeifan I think this is not a bug, but rather a QA. It has currently been changed to discussion for further questions to be discussed in the discussion.