#
# 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
import numpy as np
from transformers import AutoTokenizer, GenerationConfig


if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='Predict Tokens using `generate()` API for DeepSeek-MoE model')
    parser.add_argument('--repo-id-or-model-path', type=str, default="/mnt/disk1/models/deepseek-moe-16b-chat",
                        help='The huggingface repo id for the CodeShell model 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


    from transformers import AutoModelForCausalLM
    from ipex_llm import optimize_model
    model = AutoModelForCausalLM.from_pretrained(model_path, trust_remote_code=True, torch_dtype = torch.bfloat16, device_map = "auto", attn_implementation="eager")
    model.generation_config = GenerationConfig.from_pretrained(model_path)
    model.generation_config.pad_token_id = model.generation_config.eos_token_id
    model = optimize_model(model)
    tokenizer = AutoTokenizer.from_pretrained(model_path,
                                              trust_remote_code=True)
    
    # Generate predicted tokens
    with torch.inference_mode():
        messages = [
            {"role": "user", "content": args.prompt}
            ]
        prompt = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
        input_tensor = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
        st = time.time()
        outputs = model.generate(input_tensor.to(model.device), max_new_tokens=args.n_predict)
        end = time.time()
        result = tokenizer.decode(outputs[0][input_tensor.shape[1]:], skip_special_tokens=True)
        print(f'Inference time: {end-st} s')
        print('-'*20, 'Prompt', '-'*20)
        print(prompt)
        print('-'*20, 'Output', '-'*20)
        print(result)