Spring 2024 CS 444Assignment 3: Self-supervised and transfer learning, object detectionDue date: Wednesday, April 3, 11:59:59 PMThis assignment consists of two substantial and disjoint parts whose goals are to help you gain experience with PyTorch, learn how to use pre-trained models provided by the deep learning community, and adapt these models to new tasks and losses. In Part 1, you will use a simple self-supervised rotation prediction task to pre-train a feature representation on the CIFAR10 dataset without using class labels, and then fine-tune this representation for CIFAR10 classification. In Part 2, you will complete the implementation of an object detector based on YOLO and train it on the PASCAL dataset. Both parts will use PyTorch and Part 2 is likely to require the use of Google Colab Pro or Google Cloud Platform (GCP).Starter CodeDownload the starter code for both parts here -- and see setup instructions below under Assignment Setup. Part 1: Self-Supervised Learning by Rotation Prediction on CIFAR10Source: Gidaris et al. (2018) In Part 1, you will use PyTorch to train a model on a self-supervised task, fine-tune a subset of the model’s weights, and train a model in a fully supervised setting with different weight initializations. You will be using the CIFAR10 dataset, which is a dataset of small (32x32) images belonging to 10 different object classes. For self-supervised training, you will ignore the provided labels; however, you will use the class labels for fine-tuning and fully supervised training. The model architecture you will use is ResNet18. We will use the PyTorch ResNet18 implementation, so you do not need to create it from scratch. The self-supervised training task is image rotation prediction, as proposed by Gidaris et al. in 2018. For this task, all training images are randomly rotated by 0, 90, 180, or 270 degrees. The network is then trained to classify the rotation of each input image using cross-entropy loss by treating each of the 4 possible rotations as a class. This task can be treated as pre-training, and the pre-trained weights can then be fine-tuned on the supervised CIFAR10 classification task.
The top-level notebook (
Part 1 Extra Credit
Part 2: YOLO Object Detection on PASCAL VOCIn Part 2 you will implement a YOLO-like object detector on the PASCAL VOC 2007 dataset to produce results like in the above image.
The top-level notebook ( For full points, you should achieve an mAP close to 0.5 after 50 epochs of training. To do this, you should experiment with adjusting the batch size, but we recommend using default values for the remaining hyperparameters. As you start this part, you will realize that training is more computationally intensive than what you are used to. In order to get an idea whether your implementation works without waiting a long time for training to converge, here are reference mAP values for the first 30 epochs:
Useful ResourcesThe instructions in the The following resources are useful for understanding YOLO in detail:
Part 2 Extra Credit
Assignment SetupDownload the starter code here. To complete this assignment in a reasonable amount of time, you'll need to use a GPU. This can either be your personal GPU, Google Colab or Colab Pro with GPU enabled, or Google Cloud Platform (we will be distributing coupon codes). Environment Setup If you will be working on the assignment on a local machine then you will need a python environment set up with the appropriate packages. We suggest that you use Conda to manage python package dependencies (https://conda.io/docs/user-guide/getting-started.html). Unless you have a machine with a GPU, running this assignment on your local machine will be very slow and is not recommended. Please use Google Colab or Google Cloud Platform for this assignment. Instructions on setting up vm instances can be found here. Running Part 1 in Google Colab is fine, but a fully-trained model for Part 2 can take up to 7-8 hours. We suggest that you use Anaconda to manage python package dependencies (https://www.anaconda.com/download). This guide provides useful information on how to use Conda: https://conda.io/docs/user-guide/getting-started.html. Ensure that IPython is installed (https://ipython.org/install.html). You may then navigate the assignment directory in terminal and start a local IPython server using thejupyter notebook command.
Be careful using GOOGLE CLOUD PLATFORM!! Do not use all of your credits! Data Setup Once you have downloaded the zip file, go to the assignment3_part2 directory and execute the download_data script provided:
Submission Instructions:The assignment deliverables are as follows. If you are working in a pair, only one designated student should make the submission.
Please refer to course policies on collaborations, late submission, etc. |