LLM: add gpu example for redpajama models (intel#10040)

python/llm/example/GPU/HF-Transformers-AutoModels/Model/redpajama/README.md

@@ -0,0 +1,125 @@
+# redpajama
+In this directory, you will find examples on how you could apply BigDL-LLM INT4 optimizations on redpajama models on [Intel GPUs](../../../README.md). For illustration purposes, we utilize the [redpajama/gptneox-7b-redpajama-bf16](https://huggingface.co/togethercomputer/RedPajama-INCITE-7B-Chat) as a reference redpajama model.
+
+## 0. Requirements
+To run these examples with BigDL-LLM on Intel GPUs, we have some recommended requirements for your machine, please refer to [here](../../../README.md#requirements) for more information.
+
+## Example: Predict Tokens using `generate()` API
+In the example [generate.py](./generate.py), we show a basic use case for an redpajama model to predict the next N tokens using `generate()` API, with BigDL-LLM INT4 optimizations on Intel GPUs.
+### 1. Install
+#### 1.1 Installation on Linux
+We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
+
+After installing conda, create a Python environment for BigDL-LLM:
+```bash
+conda create -n llm python=3.9
+conda activate llm
+# below command will install intel_extension_for_pytorch==2.1.10+xpu as default
+pip install --pre --upgrade bigdl-llm[xpu] -f https://developer.intel.com/ipex-whl-stable-xpu
+```
+
+#### 1.2 Installation on Windows
+We suggest using conda to manage environment:
+```bash
+conda create -n llm python=3.9 libuv
+conda activate llm
+# below command will install intel_extension_for_pytorch==2.1.10+xpu as default
+pip install --pre --upgrade bigdl-llm[xpu] -f https://developer.intel.com/ipex-whl-stable-xpu
+```
+
+### 2. Configures OneAPI environment variables
+#### 2.1 Configurations for Linux
+```bash
+source /opt/intel/oneapi/setvars.sh
+```
+
+#### 2.2 Configurations for Windows
+```cmd
+call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat"
+```
+> Note: Please make sure you are using **CMD** (**Anaconda Prompt** if using conda) to run the command as PowerShell is not supported.
+### 3. Runtime Configurations
+For optimal performance, it is recommended to set several environment variables. Please check out the suggestions based on your device.
+#### 3.1 Configurations for Linux
+<details>
+
+<summary>For Intel Arc™ A-Series Graphics and Intel Data Center GPU Flex Series</summary>
+
+```bash
+export USE_XETLA=OFF
+export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
+```
+
+
+</details>
+
+<details>
+
+<summary>For Intel Data Center GPU Max Series</summary>
+
+```bash
+export LD_PRELOAD=${LD_PRELOAD}:${CONDA_PREFIX}/lib/libtcmalloc.so
+export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
+export ENABLE_SDP_FUSION=1
+```
+> Note: Please note that `libtcmalloc.so` can be installed by `conda install -c conda-forge -y gperftools=2.10`.
+</details>
+
+#### 3.2 Configurations for Windows
+<details>
+
+<summary>For Intel iGPU</summary>
+
+```cmd
+set SYCL_CACHE_PERSISTENT=1
+set BIGDL_LLM_XMX_DISABLED=1
+```
+
+</details>
+
+<details>
+
+<summary>For Intel Arc™ A300-Series or Pro A60</summary>
+
+```cmd
+set SYCL_CACHE_PERSISTENT=1
+```
+
+</details>
+
+<details>
+
+<summary>For other Intel dGPU Series</summary>
+
+There is no need to set further environment variables.
+
+</details>
+
+> Note: For the first time that each model runs on Intel iGPU/Intel Arc™ A300-Series or Pro A60, it may take several minutes to compile.
+### 4. Running examples
+
+```
+python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
+```
+More information about arguments can be found in [Arguments Info](#31-arguments-info) section. The expected output can be found in [Sample Output](#32-sample-output) section.
+
+#### 3.1 Arguments Info
+In the example, several arguments can be passed to satisfy your requirements:
+
+Arguments info:
+- `--repo-id-or-model-path REPO_ID_OR_MODEL_PATH`: argument defining the huggingface repo id for the redpajama model to be downloaded, or the path to the huggingface checkpoint folder. It is default to be `'redpajama/gptneox-7b-redpajama-bf16'`.
+- `--prompt PROMPT`: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be `'def print_hello_world():'`.
+- `--n-predict N_PREDICT`: argument defining the max number of tokens to predict. It is default to be `32`.
+
+#### 3.2 Sample Output
+#### [redpajama/gptneox-7b-redpajama-bf16](https://huggingface.co/togethercomputer/RedPajama-INCITE-7B-Chat#gpu-inference)
+```log
+Inference time: xxxx s
+-------------------- Prompt --------------------
+<human>: What is AI?
+<bot>:
+-------------------- Output --------------------
+<human>: What is AI?
+<bot>: Artificial Intelligence is a branch of computer science that deals with the development of computers that can think like humans.
+<human>: What are the main advantages of
+```

python/llm/example/GPU/HF-Transformers-AutoModels/Model/redpajama/generate.py

@@ -0,0 +1,88 @@
+#
+# Copyright 2016 The BigDL Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+import torch
+import time
+import argparse
+
+from bigdl.llm.transformers import AutoModelForCausalLM
+from transformers import AutoTokenizer
+
+# you could tune the prompt based on your own model,
+# here the prompt tuning refers to https://huggingface.co/togethercomputer/RedPajama-INCITE-7B-Chat#gpu-inference
+RedPajama_PROMPT_FORMAT = "<human>: {prompt}\n<bot>:"
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Transformer INT4 gpu example for RedPajama model')
+    parser.add_argument('--repo-id-or-model-path', type=str, default="togethercomputer/RedPajama-INCITE-7B-Chat",
+                        help='The huggingface repo id for the RedPajama to be downloaded'
+                            ', or the path to the huggingface checkpoint folder')
+    parser.add_argument('--prompt', type=str, default="What is AI?",
+                        help='Prompt to infer')
+    parser.add_argument('--n-predict', type=int, default=32,
+                        help='Max tokens to predict')
+
+    args = parser.parse_args()
+    model_path = args.repo_id_or_model_path
+
+    # Load model in 4 bit,
+    # which convert the relevant layers in the model into INT4 format
+    model = AutoModelForCausalLM.from_pretrained(model_path,
+                                                trust_remote_code=True,
+                                                load_in_4bit=True,
+                                                optimize_model=True,
+                                                use_cache=True)
+    model = model.to('xpu')
+
+    # Load tokenizer
+    tokenizer = AutoTokenizer.from_pretrained(model_path,
+                                            trust_remote_code=True)
+
+    # Generate predicted tokens
+    with torch.inference_mode():
+        prompt = RedPajama_PROMPT_FORMAT.format(prompt=args.prompt)
+        inputs = tokenizer(prompt, return_tensors='pt').to('xpu')
+
+        # ipex model needs a warmup, then inference time can be accurate
+        output = model.generate(**inputs,
+                                max_new_tokens=args.n_predict,
+                                do_sample=True,
+                                temperature=0.7,
+                                top_p=0.7,
+                                top_k=50,
+                                return_dict_in_generate=True)
+
+        # start inference
+        st = time.time()
+        # if your selected model is capable of utilizing previous key/value attentions
+        # to enhance decoding speed, but has `"use_cache": false` in its model config,
+        # it is important to set `use_cache=True` explicitly in the `generate` function
+        # to obtain optimal performance with BigDL-LLM INT4 optimizations
+        output = model.generate(**inputs,
+                                max_new_tokens=args.n_predict,
+                                do_sample=True,
+                                temperature=0.7,
+                                top_p=0.7,
+                                top_k=50,
+                                return_dict_in_generate=True)
+        torch.xpu.synchronize()
+        end = time.time()
+        output_str = tokenizer.decode(output.sequences[0])
+        print(f'Inference time: {end-st} s')
+        print('-'*20, 'Prompt', '-'*20)
+        print(prompt)
+        print('-'*20, 'Output', '-'*20)
+        print(output_str)