Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
yingyibiao 8f9690c4f9 | 2 years ago | |
---|---|---|
.. | ||
xlnet | 2 years ago | |
README.md | 2 years ago | |
run_all.sh | 2 years ago | |
run_benchmark.sh | 2 years ago |
目前我们为用户提供了GPT模型的Benchmark性能测试脚本。
启动测试脚本的方法如下:
ImageName="registry.baidubce.com/paddlepaddle/paddle:2.1.2-gpu-cuda10.2-cudnn7"
docker pull ${ImageName}
run_cmd="set -xe;
cd /workspace ;
bash -x tests/benchmark/run_all.sh"
nvidia-docker run --name test_paddle_gpt -i \
--net=host \
--shm-size=1g \
-v $PWD:/workspace \
${ImageName} /bin/bash -c "${run_cmd}"
如果需要打开profile选项,可以直接替换run_cmd
为:
run_cmd="set -xe;
cd /workspace ;
bash -x tests/benchmark/run_all.sh on"
黑客松task_55,在PaddleNLP的Roberta中,新增 MultipleChoice,MaskedLM 和 CausalLM三个类,7个模型权重. ,新增BPETokenizer
Python C++ Cuda Text Shell other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》