You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-[Q6: Self-Supervised Learning for Image Classification (16 points)](#q6-self-supervised-learning-16-points)
27
+
-[Optional: Style Transfer (15 points)](#optional-style-transfer-15-points)
28
28
-[Submitting your work](#submitting-your-work)
29
29
30
-
### Goals
31
-
32
-
In this assignment, you will implement recurrent neural networks and apply them to image captioning on the Microsoft COCO data. You will also explore methods for visualizing the features of a pretrained model on ImageNet, and use this model to implement Style Transfer. Finally, you will train a Generative Adversarial Network to generate images that look like a training dataset!
33
-
34
-
The goals of this assignment are as follows:
35
-
36
-
- Understand the architecture of recurrent neural networks (RNNs) and how they operate on sequences by sharing weights over time.
37
-
- Understand and implement both Vanilla RNNs and Long-Short Term Memory (LSTM) networks.
38
-
- Understand how to combine convolutional neural nets and recurrent nets to implement an image captioning system.
39
-
- Explore various applications of image gradients, including saliency maps, fooling images, class visualizations.
40
-
- Understand and implement techniques for image style transfer.
41
-
- Understand how to train and implement a Generative Adversarial Network (GAN) to produce images that resemble samples from a dataset.
42
30
43
31
### Setup
44
32
45
-
You should be able to use your setup from assignments 1 and 2.
46
-
47
-
You can work on the assignment in one of two ways: **remotely** on Google Colaboratory or **locally** on your own machine.
48
-
49
-
**Regardless of the method chosen, ensure you have followed the [setup instructions](/setup-instructions) before proceeding.**
50
-
51
-
#### Option A: Google Colaboratory (Recommended)
52
-
53
-
**Download.** Starter code containing Colab notebooks can be downloaded [here]({{site.hw_3_colab}}).
54
-
55
-
If you choose to work with Google Colab, please familiarize yourself with the [recommended workflow]({{site.baseurl}}/setup-instructions/#working-remotely-on-google-colaboratory).
33
+
Please familiarize yourself with the [recommended workflow]({{site.baseurl}}/setup-instructions/#working-remotely-on-google-colaboratory) before starting the assignment. You should also watch the Colab walkthrough tutorial below.
**Note**. Ensure you are periodically saving your notebook (`File -> Save`) so that you don't lose your progress if you step away from the assignment and the Colab VM disconnects.
60
38
61
-
Once you have completed all Colab notebooks **except `collect_submission.ipynb`**, proceed to the [submission instructions](#submitting-your-work).
62
-
63
-
#### Option B: Local Development
39
+
While we don't officially support local development, we've added a <b>requirements.txt</b> file that you can use to setup a virtual env.
64
40
65
-
**Download.** Starter code containing jupyter notebooks can be downloaded [here]({{site.hw_3_jupyter}}).
41
+
Once you have completed all Colab notebooks **except `collect_submission.ipynb`**, proceed to the [submission instructions](#submitting-your-work).
66
42
67
-
**Install Packages**. Once you have the starter code, activate your environment (the one you installed in the [Software Setup]({{site.baseurl}}/setup-instructions/) page) and run `pip install -r requirements.txt`.
43
+
### Goals
68
44
69
-
**Download data**. Next, you will need to download the COCO captioning data, a pretrained SqueezeNet model (for TensorFlow), and a few ImageNet validation images. Run the following from the `assignment3` directory:
45
+
In this assignment, you will implement recurrent neural networks and apply them to image captioning on the Microsoft COCO data. You will also explore methods for visualizing the features of a pretrained model on ImageNet, and use this model to implement Style Transfer. Finally, you will train a Generative Adversarial Network to generate images that look like a training dataset!
70
46
71
-
```bash
72
-
cd cs231n/datasets
73
-
./get_datasets.sh
74
-
```
75
-
**Start Jupyter Server**. After you've downloaded the data, you can start the Jupyter server from the `assignment3` directory by executing `jupyter notebook` in your terminal.
47
+
The goals of this assignment are as follows:
76
48
77
-
Complete each notebook, then once you are done, go to the [submission instructions](#submitting-your-work).
49
+
- Understand the architecture of recurrent neural networks (RNNs) and how they operate on sequences by sharing weights over time.
50
+
- Understand and implement Vanilla RNNs, Long-Short Term Memory (LSTM), and Transformer networks for Image captioning.
51
+
- Understand how to combine convolutional neural nets and recurrent nets to implement an image captioning system.
52
+
- Explore various applications of image gradients, including saliency maps, fooling images, class visualizations.
53
+
- Understand how to train and implement a Generative Adversarial Network (GAN) to produce images that resemble samples from a dataset.
54
+
- Understand how to leverage self-supervised learning techniques to help with image classification tasks.
55
+
-*(optional) Understand and implement techniques for image style transfer.
78
56
79
-
**You can do Questions 3, 4, and 5 in TensorFlow or PyTorch. There are two versions of each of these notebooks, one for TensorFlow and one for PyTorch. No extra credit will be awarded if you do a question in both TensorFlow and PyTorch**
57
+
**You will use PyTorch for the majority of this homework.**
80
58
81
59
### Q1: Image Captioning with Vanilla RNNs (29 points)
82
60
@@ -86,37 +64,43 @@ The notebook `RNN_Captioning.ipynb` will walk you through the implementation of
86
64
87
65
The notebook `LSTM_Captioning.ipynb` will walk you through the implementation of Long-Short Term Memory (LSTM) RNNs, and apply them to image captioning on MS-COCO.
88
66
89
-
### Q3: Network Visualization: Saliency maps, Class Visualization, and Fooling Images (15 points)
67
+
### Q3: Image Captioning with Transformers (18 points)
90
68
91
-
The notebooks `NetworkVisualization-TensorFlow.ipynb`, and `NetworkVisualization-PyTorch.ipynb` will introduce the pretrained SqueezeNet model, compute gradients with respect to images, and use them to produce saliency maps and fooling images. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awardeded if you complete both notebooks.
69
+
The notebook `Transformer_Captioning.ipynb` will walk you through the implementation of Transformer Model, and apply them to image captioning on MS-COCO.
92
70
93
-
### Q4: Style Transfer (15 points)
71
+
### Q4: Network Visualization: Saliency maps, Class Visualization, and Fooling Images (15 points)
94
72
95
-
In thenotebooks `StyleTransfer-TensorFlow.ipynb` or `StyleTransfer-PyTorch.ipynb`you will learn how to create images with the content of one image but the style of another. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awardeded if you complete both notebooks.
73
+
The notebook `NetworkVisualization-PyTorch.ipynb` will introduce the pretrained SqueezeNet model, compute gradients with respect to images, and use them to produce saliency maps and fooling images. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awardeded if you complete both notebooks.
In the notebooks `GANS-TensorFlow.ipynb` or `GANS-PyTorch.ipynb` you will learn how to generate images that match a training dataset, and use these models to improve classifier performance when training on a large amount of unlabeled data and a small amount of labeled data. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awarded if you complete both notebooks.
77
+
In the notebook `GANS-PyTorch.ipynb` you will learn how to generate images that match a training dataset, and use these models to improve classifier performance when training on a large amount of unlabeled data and a small amount of labeled data. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awarded if you complete both notebooks.
78
+
79
+
### Q6: Self-Supervised Learning (16-points)
80
+
81
+
In the notebook `self_supervised_learning,ipynb`, you will learn how to
82
+
83
+
### Optional: Style Transfer (15 points)
84
+
85
+
In thenotebooks `StyleTransfer-TensorFlow.ipynb` or `StyleTransfer-PyTorch.ipynb` you will learn how to create images with the content of one image but the style of another. Please complete only one of the notebooks (TensorFlow or PyTorch). No extra credit will be awardeded if you complete both notebooks.
100
86
101
87
### Submitting your work
102
88
103
89
**Important**. Please make sure that the submitted notebooks have been run and the cell outputs are visible.
104
90
105
-
Once you have completed all notebooks and filled out the necessary code, there are **_two_** steps you must follow to submit your assignment:
91
+
Once you have completed all notebooks and filled out the necessary code, you need to follow the below instructions to submit your work:
106
92
107
-
**1.**If you selected Option A and worked on the assignment in Colab, open `collect_submission.ipynb` in Colab and execute the notebook cells. If you selected Option B and worked on the assignment locally, run the bash script in `assignment3` by executing `bash collectSubmission.sh`.
93
+
**1.**Open `collect_submission.ipynb` in Colab and execute the notebook cells.
108
94
109
95
This notebook/script will:
110
96
111
-
* Generate a zip file of your code (`.py` and `.ipynb`) called `a3.zip`.
97
+
* Generate a zip file of your code (`.py` and `.ipynb`) called `a2.zip`.
112
98
* Convert all notebooks into a single PDF file.
113
99
114
-
**Note for Option B users**. You must have (a) `nbconvert` installed with Pandoc and Tex support and (b) `PyPDF2` installed to successfully convert your notebooks to a PDF file. Please follow these [installation instructions](https://nbconvert.readthedocs.io/en/latest/install.html#installing-nbconvert) to install (a) and run `pip install PyPDF2` to install (b). If you are, for some inexplicable reason, unable to successfully install the above dependencies, you can manually convert each jupyter notebook to HTML (`File -> Download as -> HTML (.html)`), save the HTML page as a PDF, then concatenate all the PDFs into a single PDF submission using your favorite PDF viewer.
115
-
116
100
If your submission for this step was successful, you should see the following display message:
117
101
118
-
`### Done! Please submit a3.zip and the pdfs to Gradescope. ###`
102
+
`### Done! Please submit a2.zip and the pdfs to Gradescope. ###`
119
103
120
-
**2.** Submit the PDF and the zip file to [Gradescope](https://www.gradescope.com/courses/103764).
104
+
**2.** Submit the PDF and the zip file to [Gradescope](https://www.gradescope.com/courses/257661).
121
105
122
-
**Note for Option A users**. Remember to download `a3.zip` and `assignment.pdf` locally before submitting to Gradescope.
106
+
Remember to download `a2.zip` and `assignment.pdf` locally before submitting to Gradescope.
0 commit comments