dan_scratch.py

# %%
import os


def rename_files(directory):
    for filename in os.listdir(directory):
        if filename.endswith(".csv"):
            parts = filename.split("_")
            new_name = (
                parts[2].lower().replace(" ", "_") + ".csv"
            )  # create the new filename
            os.rename(
                os.path.join(directory, filename), os.path.join(directory, new_name)
            )
            print(f"Renamed {filename} to {new_name}")


def move_files_to_subdirs(directory):
    for filename in os.listdir(directory):
        if filename.endswith(".csv"):
            subdir = os.path.join(
                directory, filename[:-4]
            )  # use the file name without .csv for the subdir name

            # Create subdirectory if it doesn't exist
            if not os.path.exists(subdir):
                os.makedirs(subdir)
                print(f"Created directory: {subdir}")

            # Move the file to the new directory
            os.rename(os.path.join(directory, filename), os.path.join(subdir, filename))
            print(f"Moved {filename} to {subdir}/{filename}")


# Specify the directory where your files are located
directory = "/Users/dan/generated_topics/"
# rename_files(directory)
move_files_to_subdirs(directory)


# %%
import os
import pandas as pd


def concat_csv_files(directory_path):
    # List all CSV files in the directory
    csv_files = [f for f in os.listdir(directory_path) if f.endswith(".csv")]

    # Read and concatenate all CSV files into a single DataFrame
    df = pd.concat(
        [pd.read_csv(os.path.join(directory_path, file)) for file in csv_files],
        ignore_index=True,
    )

    return df


# Specify the directory containing the CSV files
directory_path = "/Users/dan/generated_fixed"

# Concatenate all CSV files in the directory
concatenated_df = concat_csv_files(directory_path)

# Optionally, save the concatenated DataFrame to a new CSV file
concatenated_df.to_csv("/Users/dan/generated_dataset.csv", index=False)


# %%
#
# Load the CSV file
# base_dir = os.path.dirname(os.path.abspath(__file__))
base_dir = "/Users/dan/projects/vlm_jailbreak/prompts_and_targets/generated"
df = pd.read_csv(f"{base_dir}/generated_dataset.csv")

# Print the number of rows
print("Number of rows:", len(df))