2025-24679-hw2-q2

Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

This model performs classification on a list of images, categorizing each article of clothing by its formality.

Model Details

Model Description

<!-- Provide a longer summary of what this model is. -->

The model utilizes AutoGluon’s AutoMM for multimodal image classification. It was trained on a custom dataset of clothing images labeled as formal or informal.

Model Sources [optional]

<!-- Provide the basic links for the model. -->

• Repository: https://colab.research.google.com/drive/1uXyw9wgAI8TwPDavqMJ8OZd3HoqjDU8F?usp=sharing
• Dataset: bareethul/outfit-image-dataset

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

Detect whether a piece of clothing in an image is formal or informal.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

This was trained on a small dataset. There may be evidence of overfitting as the accuracy when evaluating the original dataset is 1.

Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Direct and downstream users should understand these limitations and risks before relying on the model in production or sensitive contexts. Further testing on broader, more diverse datasets is recommended.

How to Get Started with the Model

Use the code below to get started with the model.

MODEL_REPO_ID = "danjung9/2025-24679-hw2-q2"
zipped_predictor_filename = "autogluon_predictor_dir.zip"

# Define the local path to save the downloaded zip file
download_dir = pathlib.Path("downloaded_model")
download_dir.mkdir(exist_ok=True)
local_zip_path = download_dir / zipped_predictor_filename

# Download the zipped predictor directory from Hugging Face
huggingface_hub.hf_hub_download(
    repo_id=MODEL_REPO_ID,
    filename=zipped_predictor_filename,
    repo_type="model",
    local_dir=download_dir,
    local_dir_use_symlinks=False
)

print(f"Downloaded {zipped_predictor_filename} to {local_zip_path}")

# Unzip the predictor directory
predictor_dir_path = download_dir / "predictor"
with zipfile.ZipFile(local_zip_path, 'r') as zip_ref:
    zip_ref.extractall(predictor_dir_path.parent)

print(f"Unzipped predictor directory to {predictor_dir_path}")

# Load the predictor
loaded_predictor = autogluon.multimodal.MultiModalPredictor.load(str(predictor_dir_path))

print("Model loaded successfully!")

Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: CPU
• Hours used: 0.5
• Cloud Provider: Google
• Compute Region: US NE
• Carbon Emitted: TBD