Document and LLM dockerfile update for v2.1.30+xpu (#4181)

* update woq int4 to llm readme and llm.rst

* add doc for weight only quantization

* update version to 2.1.30

* fix doc missing files and typo

* remove gpt-j woq and add woq script

* add ipex_log to toctree and mark Prototype

* remove idex compilation and idex_dependency related

* update woq link and llm related link

* add torch ccl internal link for test

* add public torch-ccl 2.1.300

* remove llama2-34b and falcon

* fix compile issue in LLM dockerfile

* fix typo of run_bc_woq file and remove useless configuration

* add extra required packages for woq and remove code snippet

* Correct Llama 2 usage and replace PVC and ARC to official name.

* Update docker/README.md

* add ipex log link in features page

* seperate docker build for compile source and prebuilt

* sync woq int4 instruction

* fix installation command

* woq int4  link update for release branch

---------

Signed-off-by: ZhaoqiongZ <zhaoqiong.zheng@intel.com>
Co-authored-by: zhuyuhua-v <yuhua.zhu@intel.com>
Co-authored-by: Ye Ting <ting.ye@intel.com>

Author: 3 people

README.md

@@ -50,9 +50,9 @@ Compilation instruction of the latest CPU code base `main` branch can be found i
 You can install Intel® Extension for PyTorch\* for GPU via command below.
 
 ```bash
-python -m pip install torch==2.1.0a0 torchvision==0.16.0a0 torchaudio==2.1.0a0 intel-extension-for-pytorch==2.1.10+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
+python -m pip install torch==2.1.0.post2 torchvision==0.16.0.post2 torchaudio==2.1.0.post2 intel-extension-for-pytorch==2.1.30+xpu oneccl_bind_pt==2.1.300+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ 
 # for PRC user, you can check with the following link
-python -m pip install torch==2.1.0a0 torchvision==0.16.0a0 torchaudio==2.1.0a0 intel-extension-for-pytorch==2.1.10+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
+python -m pip install torch==2.1.0.post2 torchvision==0.16.0.post2 torchaudio==2.1.0.post2 intel-extension-for-pytorch==2.1.30+xpu oneccl_bind_pt==2.1.300+xpu  --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
 
 ```
 
@@ -115,3 +115,4 @@ for information on how to report a potential security issue or vulnerability.
 
 See also: [Security Policy](SECURITY.md)
 
+

dependency_version.yml

@@ -14,10 +14,12 @@ torchvision:
   commit: v0.16.0
 torch-ccl:
   repo: https://github.com/intel/torch-ccl.git
-  commit: a2164779c09bc421ad171968d5b7a9b26dcd8f0b
+  commit: 1053f1354f6293abc11e93af085524fe3664219f
   version: 2.1.300+xpu
 deepspeed:
   version: 0.14.0
+intel-extension-for-deepspeed:
+  version: 2.1.30
 transformers:
   version: 4.31.0
   commit: v4.31.0

docker/README.md

@@ -19,10 +19,10 @@ Run the following commands to build a docker image by compiling from source.
 ```
 git clone https://github.com/intel/intel-extension-for-pytorch.git
 cd intel-extension-for-pytorch
-git checkout xpu-main
+git checkout release/xpu/2.1.30
 git submodule sync
 git submodule update --init --recursive
-docker build -f docker/Dockerfile.compile --build-arg GID_RENDER=$(getent group render | sed -E 's,^render:[^:]*:([^:]*):.*$,\1,') -t intel-extension-for-pytorch:xpu .
+docker build -f docker/Dockerfile.compile --build-arg GID_RENDER=$(getent group render | sed -E 's,^render:[^:]*:([^:]*):.*$,\1,') -t intel/intel-extension-for-pytorch:2.1.30-xpu .
 ```
 
 Alternatively, `./build.sh` script has docker build command to install prebuilt wheel files, update all the relevant build arguments and execute the script. Run the command below in current directory.
@@ -98,3 +98,4 @@ Sample output looks like below:
 
 Now you are inside container with Python 3.10, PyTorch, and Intel® Extension for PyTorch\* preinstalled. You can run your own script
 to run on Intel GPU.
+

docs/images/llm/weight-only-quantization-flow.png

@@ -16,7 +16,7 @@ Once you implement and test your feature or bug-fix, submit a Pull Request to ht
 
 ## Developing Intel® Extension for PyTorch\* on XPU
 
-A full set of instructions on installing Intel® Extension for PyTorch\* from source is in the [Installation document](../../../index.html#installation?platform=gpu&version=v2.1.10%2Bxpu).
+A full set of instructions on installing Intel® Extension for PyTorch\* from source is in the [Installation document](../../../index.html#installation?platform=gpu&version=v2.1.30%2Bxpu).
 
 To develop on your machine, here are some tips:
 
@@ -126,3 +126,4 @@ To build the documentation:
 #### Tips
 
 The `.rst` source files live in `docs/tutorials` folder. Some of the `.rst` files pull in docstrings from Intel® Extension for PyTorch\* Python code (for example, via the `autofunction` or `autoclass` directives). To shorten doc build times, it is helpful to remove the files you are not working on, only keeping the base `index.rst` file and the files you are editing. The Sphinx build will produce missing file warnings but will still complete.
+

docs/images/profiler_kineto/profiler_kineto_result_console.png

@@ -178,7 +178,7 @@ For more detailed information, check `Profiler Kineto <features/profiler_kineto.
 
 
 Compute Engine (Prototype feature for debug)
------------------------------------------------
+--------------------------------------------
 
 Compute engine is a prototype feature which provides the capacity to choose specific backend for operators with multiple implementations.
 
@@ -191,3 +191,19 @@ For more detailed information, check `Compute Engine <features/compute_engine.md
    features/compute_engine
 
 
+
+
+IPEX LOG (Prototype)
+--------------------
+
+IPEX_LOGGING provides the capacity to log IPEX internal information. If you would like to use torch-style log, that is, the log/verbose is introduced by torch, and refer to cuda code, pls still use torch macro to show the log. For example, TORCH_CHECK, TORCH_ERROR. If the log is IPEX specific, or is going to trace IPEX execution, pls use IPEX_LOGGING. For some part of usage are still discussed with habana side, if has change some feature will update here.
+
+For more detailed information, check `IPEX LOG <features/ipex_log.md>`_.
+
+.. toctree::
+   :hidden:
+   :maxdepth: 1
+
+   features/ipex_log
+
+

docs/images/profiler_kineto/profiler_kineto_result_perfetto_viewer.png

@@ -10,9 +10,9 @@ Intel® Extension for PyTorch\* now empowers users to seamlessly harness graph c
 ## Required Dependencies
 
 **Verified version**:
-- `torch` : v2.1.0
+- `torch` : > v2.1.0
 - `intel_extension_for_pytorch` : > v2.1.10
-- `triton` : [v2.1.0](https://github.com/intel/intel-xpu-backend-for-triton/releases/tag/v2.1.0) with Intel® XPU Backend for Triton* backend enabled.
+- `triton` : > [v2.1.0](https://github.com/intel/intel-xpu-backend-for-triton/releases/tag/v2.1.0) with Intel® XPU Backend for Triton* backend enabled.
 
 Follow [Intel® Extension for PyTorch\* Installation](https://intel.github.io/intel-extension-for-pytorch/xpu/2.1.30+xpu/tutorials/installation.html) to install `torch` and `intel_extension_for_pytorch` firstly.
 
@@ -71,3 +71,4 @@ optimizer.zero_grad()
 loss.backward()
 optimizer.step()
 ```
+

docs/tutorials/contribution.md

@@ -1,6 +1,6 @@
 # Quick Start
 
-The following instructions assume you have installed the Intel® Extension for PyTorch\*. For installation instructions, refer to [Installation](../../../index.html#installation?platform=gpu&version=v2.1.10%2Bxpu).
+The following instructions assume you have installed the Intel® Extension for PyTorch\*. For installation instructions, refer to [Installation](../../../index.html#installation?platform=gpu&version=v2.1.30%2Bxpu).
 
 To start using the Intel® Extension for PyTorch\* in your code, you need to make the following changes:
 
@@ -58,3 +58,4 @@ source /opt/intel/oneapi/compiler/latest/env/vars.sh
 source /opt/intel/oneapi/mkl/latest/env/vars.sh
 python <script>
 ```
+