uploads.py

from email.utils import parseaddr
from huggingface_hub import HfApi
import os
import datetime
import pandas as pd


RESULTS_PATH = "boyiwei/CoTaEval_leaderboard"
api = HfApi()
TOKEN = os.environ.get("TOKEN", None)
YEAR_VERSION = "2024"

def format_error(msg):
    return f"<p style='color: red; font-size: 20px; text-align: center;'>{msg}</p>"

def format_warning(msg):
    return f"<p style='color: orange; font-size: 20px; text-align: center;'>{msg}</p>"

def format_log(msg):
    return f"<p style='color: green; font-size: 20px; text-align: center;'>{msg}</p>"

def model_hyperlink(link, model_name):
    return f'<a target="_blank" href="{link}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">{model_name}</a>'


def input_verification(model, model_family, forget_rate, url, path_to_file, organisation, mail):
    for input in [model, model_family, forget_rate, url, organisation]:
        if input == "":
            return format_warning("Please fill all the fields.")
    
    # Very basic email parsing
    _, parsed_mail = parseaddr(mail)
    if not "@" in parsed_mail:
        return format_warning("Please provide a valid email adress.")

    
    if path_to_file is None:
        return format_warning("Please attach a file.")
    
    return parsed_mail

def add_new_eval(
    model: str,
    model_family: str,
    forget_rate: str,
    url: str,
    path_to_file: str,
    organisation: str,
    mail: str,
):
    
    parsed_mail  = input_verification(model, model_family, forget_rate, url, path_to_file, organisation, mail)
    
    # load the file 
    df = pd.read_csv(path_to_file)

    # modify the df to include metadata
    df["model"] = model
    df["model_family"] = model_family
    df["forget_rate"] = forget_rate
    df["url"] = url
    df["organisation"] = organisation
    df["mail"] = parsed_mail
    df["timestamp"] = datetime.datetime.now()

    #upload to spaces using the hf api at 
    
    path_in_repo = f"versions/{model_family}-{forget_rate.replace('%', 'p')}"
    file_name = f"{model}-{organisation}-{datetime.datetime.now().strftime('%Y-%m-%d')}.csv"
    
    # upload the df to spaces
    import io
    
    buffer = io.BytesIO()
    df.to_csv(buffer, index=False)  # Write the DataFrame to a buffer in CSV format
    buffer.seek(0)  # Rewind the buffer to the beginning

    api.upload_file(
        repo_id = RESULTS_PATH,
        path_in_repo = f"{path_in_repo}/{file_name}",
        path_or_fileobj = buffer,
        token=TOKEN,
        repo_type="space",
    )

    return format_log(f"Model {model} submitted by {organisation} successfully. \nPlease refresh the leaderboard, and wait a bit to see the score displayed")