dair-ai / ml-visuals

🎨 ML Visuals contains figures and templates which you can reuse and customize to improve your scientific writing.

ML Visuals

📣 Stay tuned for significant updates to both the slides and repository.!!!

📣 In the meantime, Join our Discord

ML Visuals is a new collaborative effort to help the machine learning community in improving science communication by providing free professional, compelling and adequate visuals and figures. Currently, we have over 100 figures (all open community contributions). You are free to use the visuals in your machine learning presentations or blog posts. You don’t need to ask permission to use any of the visuals but it will be nice if you can provide credit to the designer/author (author information found in the slide notes). Check out the versions of the visuals below.

This is a project made by the dair.ai community. The latest version of the Google slides can be found in this GitHub repository. Our community members will continue to add more common figures and basic elements in upcoming versions. Think of this as free and open artifacts and templates which you can freely and easily download, copy, distribute, reuse and customize to your own needs.

ML Visuals is now being used to power 100s of figures used by master/PhD students, papers (like this one), among other use cases.

How to Use?

Essentially, we are using Google Slides to maintain all visuals and figures (check the versions below). To add your own custom figures, simply add a new slide and reuse any of the basic visual components (remember to request edit permissions). You can also create your own copy of the slides and customize whatever you like. We encourage authors/designers to add their visuals here and allow others to reuse them. Make sure to include your author information (in the notes section of the slide) so that others can provide credit if they use the visuals elsewhere (e.g. blog/presentations). Also, provide a short description of your visual to help the user understand what it is about and how they can use it. If you need "Edit" permission, just click on the "request edit access" option under the "view only" toolbar (in Google Slides) or send me an email at [email protected].

Downloading a figure from any of the slides is easy. Just click on File→Download→(choose your format).

If you need help with customizing a figure or have an idea of something that could be valuable to others, we can help. Just open an issue here and we will do our best to come up with the visual. Thanks.

Feel free to reach out to me on Twitter for an invite to our Slack group.

Versions:

• Version 1.0

How to Contribute?

• You can check out our Project page to see all the ongoing tasks or issues related to this research project. Lookout for the main ml_visuals tag. Issues with the good first issue tag are good tasks to get started with.
• You can also just check the issues tab.
• You can ask anything related to this project in our Slack group
• Slack channel: #ml_visuals

Some ideas for figures to add to the Slides (issue)

• Linear regression, single-layer neural network
• Multilayer Perceptron with hidden layer
• Backpropagation
• Batch Normalization and alternatives
• Computational Graphs
• Dropout
• CNN - padding, stride, pooling,...
• LeNet
• AlexNet
• VGG
• GoogleNet
• ResNet
• DenseNet
• Memory Networks
• RNN
• Deep RNN
• Bidirectional RNN
• GRU
• LSTM
• Language RNN models
• Backpropagation through time
• Encoder-Decoder Architecture
• Seq2seq with RNN encoder-decoder
• Bearm search and other decoding strategies
• Attention
• Multi-head attention
• Self-attention
• Transformer
• Word2vec/GloVe/Skip-gram/CBOW/BERT/GPT....
• Common/Popular CV/NLP Tasks

List adopted from multiple resources including nlpoverview and d2l.ai which both contain a very solid syllabus.

Examples of Visuals

ml-visuals's Issues

女

CBAM

cnn

Add classical/new architectures

Some ideas for figures to add to the PPT

• Linear regression, single-layer neural network
• Multilayer Perceptron with hidden layer
• Backpropagation
• Batch Normalization and alternatives
• Computational Graphs
• Dropout
• CNN - padding, stride, pooling,...
• LeNet
• AlexNet
• VGG
• GoogleNet
• ResNet
• DenseNet
• Memory Networks
• RNN
• Deep RNN
• Bidirectional RNN
• GRU
• LSTM
• Language RNN models
• Backpropagation through time
• Encoder-Decoder Architecture
• Seq2seq with RNN encoder-decoder
• Bearm search and other decoding strategies
• Attention
• Multi-head attention
• Self-attention
• Transformer
• Word2vec/GloVe/Skip-gram/CBOW/BERT/GPT....
• Common/Popular CV/NLP Tasks

List adopted from multiple resources including nlpoverview and d2l.ai which both contain a very solid syllabus.

Please feel free to make suggestions below. If you would like to help, also let me know.

minshan

name: "PointCloudModel"
layer {
name: "data"
type: "Input"
top: "data"
input_param { shape: { dim: 1 dim: 9 dim: 64 } }
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
convolution_param {
num_output: 64
kernel_size: 1
kernel_size: 9
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv1"
top: "conv2"
convolution_param {
num_output: 64
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv2"
top: "conv3"
convolution_param {
num_output: 64
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
convolution_param {
num_output: 128
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
convolution_param {
num_output: 1024
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "maxpool1"
type: "Pooling"
bottom: "conv5"
top: "maxpool1"
pooling_param {
pool: MAX
kernel_size: 1
kernel_size: 64
}
}
layer {
name: "fc1"
type: "InnerProduct"
bottom: "maxpool1"
top: "fc1"
inner_product_param {
num_output: 256
}
}
layer {
name: "fc2"
type: "InnerProduct"
bottom: "fc1"
top: "fc2"
inner_product_param {
num_output: 128
}
}
layer {
name: "conv6"
type: "Convolution"
bottom: "conv5"
top: "conv6"
convolution_param {
num_output: 512
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv6"
top: "conv7"
convolution_param {
num_output: 256
kernel_size: 1
kernel_size: 1
}
}
layer {
name: "conv8"
type: "Convolution"
bottom: "conv7"
top: "conv8"
convolution_param {
num_output: 13
kernel_size: 1
kernel_size: 1
}
}

内容是空的，没有版权了？？？

GEE

DNN

I can't find the download link

Dear authors, I have download the project, but i don't know where is the ppt.
Thank you.

Citation

Glad to find these excellent visualizations just in time as I started to write my thesis. Was wondering how can I cite this work in my thesis/paper later on?

permission request

I clicked the permission get button but no one seems to be responding

Encoder

pronlem

can't find document of PPT

CRNN

could not open the downloaded file

Hi, team, I tried to download the file as pptx format, but when it finished, my microsoft power point 2019 can not open it. The PPT2019 notified that the file is broken and needs to be recovered. But actually the recovery process cannot make it readable either... Anyone knows which app is able to open the slide file?

版权问题

如果在SCI论文中使用了贵方的图像，需要进行引用吗？引用格式包括图像来源和引用文献的作者、标题、期刊名称、年份和页码等信息可以提供吗？小白发问，勿喷

Can we add federated learning frameworks?

This is a fantastic repo and thanks for the contribution to the community! Can we add some framework graph for federated learning?

How to draw the graph on page 31？

Hello author, I checked and studied your project, but I still don’t know how to draw the other line chart in google ppt on page 31? Or can you tell me what tools are used?

ad

Can we add Graph Neural Network models?

This is a fantastic repo. Can we Graph Neural Network models like GCN, GAT, GGNN, GraphSage, etc?

cnn

question

how to use

Add your name to the contributors list

Please say hi and add your name below if you wish to contribute to this project. Make sure to link your GitHub account so that I can add you to this project.

eeg emotion recognition

bert

ee

erw

折线图

绘制一张折线图

scidraw.io

Hi, what a cool initiative, nice done!

We had something similar in mind when we created scidraw.io, though with a more biology focused perspective.

Though we are starting to see some ml related content, it's not enough. If you want to also share your content on scidraw to reach a wider audience, get in touch :)

Translate README.md to Japanese.

I translated it into Japanese, so you can use it if you like. I hope you like it. (I just wrote it in markdown format.)

ML ビジュアル

ML Visuals は、専門的で説得力のある適切なビジュアルや図形を無料で提供することで、機械学習コミュニティが科学コミュニケーションを改善するのを支援するための新しい共同作業です。現在、100以上の図があります（すべてオープンなコミュニティの貢献です）。あなたの機械学習のプレゼンテーションやブログ記事でビジュアルを自由に使用することができます。ビジュアルの使用許可を求める必要はありませんが、デザイナー/著者へのクレジット（スライドノートに記載されている著者情報）を提供していただければ幸いです。以下のビジュアルのバージョンをチェックしてください。

これはdair.aiコミュニティによって作られたプロジェクトです。Googleのスライドの最新バージョンは、このGitHubリポジトリにあります。私たちのコミュニティメンバーは、今後のバージョンでより一般的な数字や基本的な要素を追加していきます。これは自由でオープンな成果物やテンプレートであり、自由に簡単にダウンロードしたり、コピーしたり、配布したり、再利用したり、自分のニーズに合わせてカスタマイズしたりすることができると考えてください。

ML Visuals は現在，修士・博士課程の学生や論文（今回のようなもの）などで使用されている 100 種類以上の図の動力源として使用されています．

どのように使用するのか？

基本的には、すべてのビジュアルと図を維持するためにGoogle Slidesを使用しています（以下のバージョンを確認してください）。独自のカスタム図を追加するには、単に新しいスライドを追加し、基本的なビジュアルコンポーネントのいずれかを再利用します（編集権限を要求することを忘れないでください）。また、スライドのコピーを作成して、好きなようにカスタマイズすることもできます。私たちは、著者/デザイナーがここに彼らのビジュアルを追加し、他の人がそれらを再利用できるようにすることをお勧めします。他の人が他の場所（例えば、ブログ/プレゼンテーション）でビジュアルを使用した場合にクレジットを提供できるように、あなたの著者情報（スライドのノートセクションに）を含めることを確認してください。また、ユーザーがそれがどのようなもので、どのように使用できるのかを理解するのに役立つように、ビジュアルの簡単な説明を記載してください。編集」の許可が必要な場合は、「表示専用」ツールバー（Google Slides）の下にある「編集アクセスを要求する」オプションをクリックするか、[email protected] までメールでご連絡ください。

スライドのいずれかから図をダウンロードするのは簡単です。ファイル→ダウンロード→（フォーマットを選択）をクリックするだけです。

図をカスタマイズする際に助けが必要な場合や、他の人にとって価値のあるものを考えている場合は、私たちがお手伝いします。ここ で課題を開いていただければ、私たちは最善を尽くしてビジュアルを考え出します。ありがとうございます。

Slackグループへの招待については、 Twitterで私に連絡してください。

バージョン

• Version 1.0

貢献するには？

この研究プロジェクトに関連した進行中のタスクや問題を見るために、 プロジェクトページをチェックすることができます。メインの ml_visuals タグを探してください。良い最初の課題タグを持つ課題は、始めるのに良いタスクです。
また、課題タブをチェックすることもできます。
このプロジェクトに関連することは何でも私たちの Slack グループで質問することができます。
Slackチャンネル。#ml_visuals

ビジュアライズの例

![1.png]

![2.png]

![3.png]

Can we add contrastive representation learning models?

Thanks for this awesome repo. This helps a lot.

Wondering can we also including Contrastive Representation Learning models?

cnn

anomaly detection

ff

fff

stack

there are some different graph which were added by someone

I found that int the version 1.0 there are some different graph which were added by someone, they asked the permission to edit but they didn't know that their ppt can be seen by everyone. They should download it and edit it on their personal devices.

编码器-解码器架构

编码器-解码器架构

情绪识别

脑电地形图，功率谱图，分类器可视化效果，各种脑电地形图

q

Other download links available？

Many users in China cannot access Google Slide, can you provide other download links? Thanks.

attention

你好，你这个用不了啊，工程里没有什么东西呀

CNN