search

intel-analytics / ipex-llm

Accelerate local LLM inference and finetuning (LLaMA, Mistral, ChatGLM, Qwen, Baichuan, Mixtral, Gemma, Phi, etc.) on Intel CPU and GPU (e.g., local PC with iGPU, discrete GPU such as Arc, Flex and Max); seamlessly integrate with llama.cpp, Ollama, HuggingFace, LangChain, LlamaIndex, DeepSpeed, vLLM, FastChat, Axolotl, etc.
Apache License 2.0
6.29k stars 1.23k forks source link

cannot setup chronos from source #4988

Open onurbarut opened 2 years ago

onurbarut commented 2 years ago

When I installed bigdl-chronos using this guide and want to use autoformer forcaster, TSDataset complains about not having 'time_enc' as one of arguments. when I check bigdl-version, it is 2.0.0. Then, I wanted to install chronos from source, but while installing I got error bigdl-orca==2.1.0.dev0 not found to be installed (see below).

... creating 'dist/bigdl_chronos-2.1.0.dev0-py3.7.egg' and adding 'build/bdist.linux-x86_64/egg' to it removing 'build/bdist.linux-x86_64/egg' (and everything under it) Processing bigdl_chronos-2.1.0.dev0-py3.7.egg removing '/home/onur/intel/oneapi/intelpython/python3.9/envs/autots2/lib/python3.7/site-packages/bigdl_chronos-2.1.0.dev0-py3.7.egg' (and everything under it) creating /home/onur/intel/oneapi/intelpython/python3.9/envs/autots2/lib/python3.7/site-packages/bigdl_chronos-2.1.0.dev0-py3.7.egg Extracting bigdl_chronos-2.1.0.dev0-py3.7.egg to /home/onur/intel/oneapi/intelpython/python3.9/envs/autots2/lib/python3.7/site-packages bigdl-chronos 2.1.0.dev0 is already the active version in easy-install.pth

Installed /home/onur/intel/oneapi/intelpython/python3.9/envs/autots2/lib/python3.7/site-packages/bigdl_chronos-2.1.0.dev0-py3.7.egg Processing dependencies for bigdl-chronos==2.1.0.dev0 Searching for bigdl-orca==2.1.0.dev0 Reading https://pypi.org/simple/bigdl-orca/ No local packages or working download links found for bigdl-orca==2.1.0.dev0 error: Could not find suitable distribution for Requirement.parse('bigdl-orca==2.1.0.dev0')

Thanks for help!

TheaperDeng commented 2 years ago

Hi, thanks for your feedback.

stable version bigdl-chronos(2.0.0) does not have autoformer model or related function in TSDataset, but the nightly-built version does.

Please install nightly-build version bigdl-chronos through

pip install --pre --upgrade bigdl-chronos[all]

Detailed information can be found here: https://bigdl.readthedocs.io/en/latest/doc/Chronos/Overview/chronos.html#install

Since our nightly-build version will be released every day, you don't need to build from src.

TheaperDeng commented 2 years ago

Since autoformer is a very new model to chronos, some updates/bug fixes might be done in next few weeks.

A good example to start with autoformer is: https://github.com/intel-analytics/BigDL/tree/main/python/chronos/use-case/electricity