MonteXiaofeng/Tranport-llama3_1_8B_instruct

This model is finetuned on the model llama3.1-8b-instruct using the dataset BAAI/IndustryInstruction_Transportation dataset, the dataset details can jump to the repo: BAAI/IndustryInstruction

training params

The training framework is llama-factory, template=llama3

learning_rate=1e-5
lr_scheduler_type=cosine
max_length=2048
warmup_ratio=0.05
batch_size=64
epoch=10

select best ckpt by the evaluation loss

evaluation

Since I only found an instruction dataset DUOMO-Lab/Transgpt_sft_v2 in the field of traffic, in order to remove the influence of the base model, I used the data in llama3.1-8b-instruc for fine-tuning and compared and evaluated our model. The evaluation method is: use GPT4 on the validation set of each dataset to compare good, tie, and loss. The evaluation results are as follows

How to use

# !/usr/bin/env python
# -*- coding:utf-8 -*-
# ==================================================================
# [Author]       : xiaofeng
# [Descriptions] :
# ==================================================================

from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch


llama3_jinja = """{% if messages[0]['role'] == 'system' %}
    {% set offset = 1 %}
{% else %}
    {% set offset = 0 %}
{% endif %}

{{ bos_token }}
{% for message in messages %}
    {% if (message['role'] == 'user') != (loop.index0 % 2 == offset) %}
        {{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}
    {% endif %}

    {{ '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n' + message['content'] | trim + '<|eot_id|>' }}
{% endfor %}

{% if add_generation_prompt %}
    {{ '<|start_header_id|>' + 'assistant' + '<|end_header_id|>\n\n' }}
{% endif %}"""


dtype = torch.bfloat16

model_dir = "MonteXiaofeng/Tranport-llama3_1_8B_instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_dir,
    device_map="cuda",
    torch_dtype=dtype,
)

tokenizer = AutoTokenizer.from_pretrained(model_dir)
tokenizer.chat_template = llama3_jinja # update template

message = [
    {"role": "system", "content": "You are a helpful assistant"},
    {"role": "user", "content": "私人交通工具的发展对经济有什么影响？"},
]
prompt = tokenizer.apply_chat_template(
    message, tokenize=False, add_generation_prompt=True
)
print(prompt)
inputs = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt")
prompt_length = len(inputs[0])
print(f"prompt_length:{prompt_length}")

generating_args = {
    "do_sample": True,
    "temperature": 1.0,
    "top_p": 0.5,
    "top_k": 15,
    "max_new_tokens": 512,
}


generate_output = model.generate(input_ids=inputs.to(model.device), **generating_args)

response_ids = generate_output[:, prompt_length:]
response = tokenizer.batch_decode(
    response_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True
)
print(response)
"""
私人交通工具的发展对经济有着深远的影响。首先，私人交通工具的发展可以促进汽车制造业的繁荣。随着私人交通工具的需求增加，汽车制造商将面临更大的市场需求，从而带动产业链的发展，创造就业机会，增加经济收入。其次，私人交通工具的发展也会带动相关
  业的发展，如燃料供应、维修服务和保险等。这些行业的发展将为经济增长做出贡献。此外，私人交通工具的发展还会促进城市交通的便利性，提高人们的生活质量，从而带动消费，刺激经济发展。然而，私人交通工具的发展也会带来一些负面影响，如交通拥堵和环境
  染等问题。因此，政府需要采取相应的政策措施来平衡经济发展和环境保护的需要。总的来说，私人交通工具的发展对经济有着重要的影响，需要综合考虑各种因素进行合理规划和管理。
"""