vjepa-self-driving

VJEPA Encoder

The VJEPA Encoder is a Python package that provides an implementation of the encoder component from the JEPA (Joint Encoding for Prediction and Alignment) architecture proposed by Facebook AI Research. The encoder is designed to extract meaningful representations from visual data. I do not own the rights or lay claim to the copyright of this software. This package is an adaptation to facebookresearch/jepa to enable ease of use of the Jepa Architecture built with Vision Transformers.

Installation

To install the VJEPA Encoder package, you can use pip:

pip install vjepa-encoder

Usage

To use the VJEPA Encoder in your Python code, you can import it as follows:

from vjepa_encoder.vision_encoder import JepaEncoder

Loading the Encoder

To load the pre-trained encoder, you can use the load_model function:

config_file_path = "./params-encoder.yaml"
devices = ["cuda:0"]
encoder = JepaEncoder.load_model(config_file_path, devices)

• config_file_path: Path to the configuration file (YAML) containing the model settings.
• devices: List of devices (e.g., ['cuda:0']) to use for distributed training. If not provided, the model will be loaded on the CPU.

Important Notes about the Config File:

• the config file provided in this repo provides the basics for loading and using the encoder. The most important things to note in this file are the r_checkpoint: points at the .tar file for the JEPA checkpoint, and the tabulet_size: this is used in some temporal calculation and if you plan on embedding images you should set this to 1; set this to N if you plan on using a temporal dimension for your data, where N corresponds to however many temporal inputs you have.

Preprocessing Data

The VJEPA Encoder provides a preprocess_data function to preprocess input data before feeding it to the encoder:

preprocessed_data = encoder.preprocess_data(input_data)

• input_data: Input data, which can be an image path, image array, PIL Image, or PyTorch tensor.

Embedding Images

To obtain the embeddings for an image, you can use the embed_image function:

embeddings = encoder.embed_image(input_data)

• input_data: Input data, which can be an image path, image array, PIL Image, or PyTorch tensor.

The function returns the embeddings generated by the encoder.

Configuration

The VJEPA Encoder requires a configuration file in YAML format to specify the model settings. The configuration file should include the following sections:

• meta: General settings such as the checkpoint file path, random seed, etc.
• mask: Settings related to masking.
• model: Model architecture settings.
• data: Data-related settings such as crop size, patch size, etc.
• logging: Logging settings.

Please refer to the provided configuration file template for more details.

License

The VJEPA Encoder is released under the MIT License.

Acknowledgments

The VJEPA Encoder is based on the research work conducted by Facebook AI Research. We would like to acknowledge their contributions to the field of computer vision and representation learning.

Contact

If you have any questions or suggestions regarding the VJEPA Encoder, please feel free to contact us at [email protected].

---