utils/load_csv.py

from dotenv import load_dotenv
from huggingface_hub import HfApi, hf_hub_download
import os
import io
import pandas as pd

# Load environment variables from .env file
load_dotenv()

# ASR_model = "openai/whisper-largev2"  # Replace with your ASR model
# csv_path = "test.csv" #read from local
# csv_transcript = f"test_with_{ASR_model.replace("/","_")}.csv" # to save in dataset repo
# csv_result = f"test_with_{ASR_model.replace("/","_")}_WER.csv" # to save in dataset repo

# df = pd.read_csv(csv_path)
# print(f"CSV Loaded with {len(df)} rows")

def upload_csv(df,csv_filename):
    csv_buffer = io.BytesIO()
    df.to_csv(csv_buffer, index=False)
    csv_buffer.seek(0)

    try:
        # Upload the generated csv to Hugging Face Hub
        api = HfApi(token=os.getenv("HF_TOKEN"))
        print(f"✅ CSV uploading : {csv_filename}")
        api.upload_file(
            path_or_fileobj=csv_buffer,
            path_in_repo=csv_filename,
            repo_id="satyamr196/asr_fairness_results",
            repo_type="dataset"
        )
        return True
    except Exception as e:
        print(f"⚠️ Could not upload CSV: {csv_filename} — {e}")
        return False

# upload_csv(df,f"test_with_{ASR_model.replace("/","_")}_WER.csv");

def download_csv(csv_filename):
    repo_id = "satyamr196/asr_fairness_results"

    try:
        # Download the CSV file from the dataset repo
        csv_path = hf_hub_download(repo_id=repo_id, filename=csv_filename, repo_type="dataset")
        # Load into pandas
        return pd.read_csv(csv_path)
    except Exception as e:
        # print(f"⚠️ Could not load CSV: {csv_filename} — {e}")
        return None


# # # Load the csv from the Hugging Face Hub
# df = download_csv(csv_result)
# if(df is None):
#     print(f"CSV not found in the dataset repo. Please upload the file first.")
# else:
#     print(f"CSV Loaded with {len(df)} rows")

# print(df)