# SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# 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 argparse
import os
import time

import tensorrt_llm
from tensorrt_llm.models import Phi3ForCausalLM, PhiForCausalLM


def parse_arguments():
    parser = argparse.ArgumentParser()
    parser.add_argument('--model_dir', type=str, default=None)
    parser.add_argument('--dtype',
                        type=str,
                        default='float16',
                        choices=['float32', 'bfloat16', 'float16'])
    parser.add_argument('--output_dir',
                        type=str,
                        default='tllm_checkpoint',
                        help='The path to save the TensorRT-LLM checkpoint')
    parser.add_argument(
        '--model_type',
        type=str,
        default='phi-2',
        choices=['phi-2', 'Phi-3-mini-4k-instruct', 'Phi-3-mini-128k-instruct'],
        help='Model to be converted.')
    args = parser.parse_args()

    return args


if __name__ == '__main__':
    print(tensorrt_llm.__version__)
    args = parse_arguments()

    tik = time.time()
    if not os.path.exists(args.output_dir):
        os.makedirs(args.output_dir)

    modelForCausalLM = PhiForCausalLM if args.model_type == "phi-2" else Phi3ForCausalLM
    modelForCausalLM.convert_hf_checkpoint(args.model_dir,
                                           dtype=args.dtype,
                                           output_dir=args.output_dir)

    tok = time.time()
    t = time.strftime('%H:%M:%S', time.gmtime(tok - tik))
    print(f'Total time of converting checkpoints: {t}')