app.py

import gradio as gr
from alignment import (
    DataArguments,
    ModelArguments,
    apply_chat_template,
    get_datasets,
    get_tokenizer,
)

def template(base_model, trained_adapter, token):
    data_args = DataArguments(chat_template=None, dataset_mixer={'HuggingFaceH4/no_robots': 1.0}, dataset_splits=['train_sft', 'test_sft'], max_train_samples=None, max_eval_samples=None, preprocessing_num_workers=12, truncation_side=None)
    model_args = ModelArguments(base_model_revision=None, model_name_or_path='mistralai/Mistral-7B-v0.1', model_revision='main', model_code_revision=None, torch_dtype='auto', trust_remote_code=True, use_flash_attention_2=True, use_peft=True, lora_r=64, lora_alpha=16, lora_dropout=0.1, lora_target_modules=['q_proj', 'k_proj', 'v_proj', 'o_proj'], lora_modules_to_save=None, load_in_8bit=False, load_in_4bit=True, bnb_4bit_quant_type='nf4', use_bnb_nested_quant=False)
    
    ###############
    # Load datasets
    ###############
    raw_datasets = get_datasets(data_args, splits=data_args.dataset_splits)
    logger.info(
        f"Training on the following datasets and their proportions: {[split + ' : ' + str(dset.num_rows) for split, dset in raw_datasets.items()]}"
    )
    
    ################
    # Load tokenizer
    ################
    tokenizer = get_tokenizer(model_args, data_args)
    
    #####################
    # Apply chat template
    #####################
    raw_datasets = raw_datasets.map(apply_chat_template, fn_kwargs={"tokenizer": tokenizer, "task": "sft"})
    train_dataset = raw_datasets["train"]
    eval_dataset = raw_datasets["test"]

with gr.Blocks() as demo:
    gr.Markdown("## AutoTrain Merge Adapter")
    gr.Markdown("Please duplicate this space and attach a GPU in order to use it.")
    token = gr.Textbox(
        label="Hugging Face Write Token",
        value="",
        lines=1,
        max_lines=1,
        interactive=True,
        type="password",
    )
    base_model = gr.Textbox(
        label="Base Model (e.g. meta-llama/Llama-2-7b-chat-hf)",
        value="",
        lines=1,
        max_lines=1,
        interactive=True,
    )
    trained_adapter = gr.Textbox(
        label="Trained Adapter Model (e.g. username/autotrain-my-llama)",
        value="",
        lines=1,
        max_lines=1,
        interactive=True,
    )
    submit = gr.Button(value="Merge & Push")
    op = gr.Markdown(interactive=False)
    submit.click(merge, inputs=[base_model, trained_adapter, token], outputs=[op])


if __name__ == "__main__":
    demo.launch()