5079ed9e06
* Add Llama3.1 example Add Llama3.1 example for Linux arc and Windows MTL * Changes made to adjust compatibilities transformers changed to 4.43.1 * Update index.rst * Update README.md * Update index.rst * Update index.rst * Update index.rst
81 lines
3.6 KiB
Python
81 lines
3.6 KiB
Python
#
|
|
# 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 ipex_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://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1
|
|
DEFAULT_SYSTEM_PROMPT = """\
|
|
"""
|
|
|
|
def get_prompt(user_input: str, chat_history: list[tuple[str, str]],
|
|
system_prompt: str) -> str:
|
|
prompt_texts = [f'<|begin_of_text|>']
|
|
|
|
if system_prompt != '':
|
|
prompt_texts.append(f'<|start_header_id|>system<|end_header_id|>\n\n{system_prompt}<|eot_id|>')
|
|
|
|
for history_input, history_response in chat_history:
|
|
prompt_texts.append(f'<|start_header_id|>user<|end_header_id|>\n\n{history_input.strip()}<|eot_id|>')
|
|
prompt_texts.append(f'<|start_header_id|>assistant<|end_header_id|>\n\n{history_response.strip()}<|eot_id|>')
|
|
|
|
prompt_texts.append(f'<|start_header_id|>user<|end_header_id|>\n\n{user_input.strip()}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n')
|
|
return ''.join(prompt_texts)
|
|
|
|
if __name__ == '__main__':
|
|
parser = argparse.ArgumentParser(description='Predict Tokens using `generate()` API for Llama3.1 model')
|
|
parser.add_argument('--repo-id-or-model-path', type=str, default="meta-llama/Meta-Llama-3.1-8B-Instruct",
|
|
help='The huggingface repo id for the Llama3 (e.g. `meta-llama/Meta-Llama-3.1-8B-Instruct`) 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,
|
|
load_in_4bit=True,
|
|
optimize_model=True,
|
|
trust_remote_code=True,
|
|
use_cache=True)
|
|
|
|
# Load tokenizer
|
|
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
|
|
|
|
# Generate predicted tokens
|
|
with torch.inference_mode():
|
|
prompt = get_prompt(args.prompt, [], system_prompt=DEFAULT_SYSTEM_PROMPT)
|
|
input_ids = tokenizer.encode(prompt, return_tensors="pt")
|
|
st = time.time()
|
|
output = model.generate(input_ids,
|
|
max_new_tokens=args.n_predict)
|
|
end = time.time()
|
|
output_str = tokenizer.decode(output[0], skip_special_tokens=False)
|
|
print(f'Inference time: {end-st} s')
|
|
print('-'*20, 'Prompt', '-'*20)
|
|
print(prompt)
|
|
print('-'*20, 'Output (skip_special_tokens=False)', '-'*20)
|
|
print(output_str)
|
|
|