├── LICENSE ├── README.md └── docs ├── README.md ├── _sidebar.md ├── chapter1 ├── images │ ├── CH1-image1.PNG │ ├── CH1-image10.png │ ├── CH1-image11.png │ ├── CH1-image2.png │ ├── CH1-image3.png │ ├── CH1-image4.jpg │ ├── CH1-image5.png │ ├── CH1-image6.png │ ├── CH1-image7.png │ ├── CH1-image8.png │ ├── CH1-image9.png │ ├── gru-3.svg │ ├── gru_1.svg │ ├── lstm-3.svg │ ├── lstm_1.svg │ ├── rnn.svg │ └── rnn_1.svg ├── introduction.md ├── task1.md ├── 低维映射到高维.ipynb └── 循环神经网络.md ├── chapter2 ├── Transformer.md ├── WordVec.ipynb ├── hongloumeng.txt ├── images │ ├── C2images1.png │ ├── C2images10.png │ ├── C2images11.png │ ├── C2images12.png │ ├── C2images13.png │ ├── C2images14.png │ ├── C2images2.png │ ├── C2images3.png │ ├── C2images4.png │ ├── C2images5.png │ ├── C2images6.png │ ├── C2images7.png │ ├── C2images8.png │ ├── C2images9.png │ └── wordvec.png ├── sanguoyanyi.txt ├── task2.md ├── word2vec_gensim ├── word2vec_gensim1 ├── word2vec_gensim2 ├── 相对位置向量.ipynb ├── 词向量生成过程.md └── 软注意力和硬注意力.md ├── chapter3 ├── CrossAttention.ipynb ├── Encoder.md ├── images │ ├── C3image1.PNG │ ├── C3image10.png │ ├── C3image2.PNG │ ├── C3image3.PNG │ ├── C3image4.png │ ├── C3image5.png │ ├── C3image6.png │ ├── C3image7.png │ ├── C3image8.png │ ├── C3image9.png │ ├── C3images11.png │ └── C3images12.png └── task3.md ├── chapter4 ├── Decoder.md ├── Tokenization.md ├── apply-bert.ipynb ├── apply-gpt.ipynb ├── images │ ├── C4images1.PNG │ ├── C4images2.png │ ├── C4images3.png │ ├── Token1.png │ └── Token2.png └── task4.md ├── chapter5 ├── Transformer组件实现.ipynb ├── project.md ├── task5.md ├── test.py ├── 使用 NumPy 和 SciPy 实现通用注意力机制.ipynb ├── 多头注意力机制.ipynb ├── 实践项目.ipynb ├── 实践项目_GPU.ipynb └── 自注意力机制实现.ipynb └── chapter6 ├── README.md └── Transformer_for_vision.md /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/README.md -------------------------------------------------------------------------------- /docs/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/README.md -------------------------------------------------------------------------------- /docs/_sidebar.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/_sidebar.md -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image1.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image1.PNG -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image10.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image11.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image11.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image2.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image3.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image4.jpg -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image5.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image6.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image7.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image8.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image8.png -------------------------------------------------------------------------------- /docs/chapter1/images/CH1-image9.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/CH1-image9.png -------------------------------------------------------------------------------- /docs/chapter1/images/gru-3.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/gru-3.svg -------------------------------------------------------------------------------- /docs/chapter1/images/gru_1.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/gru_1.svg -------------------------------------------------------------------------------- /docs/chapter1/images/lstm-3.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/lstm-3.svg -------------------------------------------------------------------------------- /docs/chapter1/images/lstm_1.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/lstm_1.svg -------------------------------------------------------------------------------- /docs/chapter1/images/rnn.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/rnn.svg -------------------------------------------------------------------------------- /docs/chapter1/images/rnn_1.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/images/rnn_1.svg -------------------------------------------------------------------------------- /docs/chapter1/introduction.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/introduction.md -------------------------------------------------------------------------------- /docs/chapter1/task1.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/task1.md -------------------------------------------------------------------------------- /docs/chapter1/低维映射到高维.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/低维映射到高维.ipynb -------------------------------------------------------------------------------- /docs/chapter1/循环神经网络.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter1/循环神经网络.md -------------------------------------------------------------------------------- /docs/chapter2/Transformer.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/Transformer.md -------------------------------------------------------------------------------- /docs/chapter2/WordVec.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/WordVec.ipynb -------------------------------------------------------------------------------- /docs/chapter2/hongloumeng.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/hongloumeng.txt -------------------------------------------------------------------------------- /docs/chapter2/images/C2images1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images1.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images10.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images11.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images11.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images12.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images12.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images13.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images13.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images14.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images14.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images2.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images3.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images4.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images5.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images6.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images7.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images8.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images8.png -------------------------------------------------------------------------------- /docs/chapter2/images/C2images9.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/C2images9.png -------------------------------------------------------------------------------- /docs/chapter2/images/wordvec.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/images/wordvec.png -------------------------------------------------------------------------------- /docs/chapter2/sanguoyanyi.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/sanguoyanyi.txt -------------------------------------------------------------------------------- /docs/chapter2/task2.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/task2.md -------------------------------------------------------------------------------- /docs/chapter2/word2vec_gensim: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/word2vec_gensim -------------------------------------------------------------------------------- /docs/chapter2/word2vec_gensim1: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/word2vec_gensim1 -------------------------------------------------------------------------------- /docs/chapter2/word2vec_gensim2: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/word2vec_gensim2 -------------------------------------------------------------------------------- /docs/chapter2/相对位置向量.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/相对位置向量.ipynb -------------------------------------------------------------------------------- /docs/chapter2/词向量生成过程.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/词向量生成过程.md -------------------------------------------------------------------------------- /docs/chapter2/软注意力和硬注意力.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter2/软注意力和硬注意力.md -------------------------------------------------------------------------------- /docs/chapter3/CrossAttention.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/CrossAttention.ipynb -------------------------------------------------------------------------------- /docs/chapter3/Encoder.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/Encoder.md -------------------------------------------------------------------------------- /docs/chapter3/images/C3image1.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image1.PNG -------------------------------------------------------------------------------- /docs/chapter3/images/C3image10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image10.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image2.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image2.PNG -------------------------------------------------------------------------------- /docs/chapter3/images/C3image3.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image3.PNG -------------------------------------------------------------------------------- /docs/chapter3/images/C3image4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image4.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image5.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image6.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image7.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image8.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image8.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3image9.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3image9.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3images11.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3images11.png -------------------------------------------------------------------------------- /docs/chapter3/images/C3images12.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/images/C3images12.png -------------------------------------------------------------------------------- /docs/chapter3/task3.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter3/task3.md -------------------------------------------------------------------------------- /docs/chapter4/Decoder.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/Decoder.md -------------------------------------------------------------------------------- /docs/chapter4/Tokenization.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/Tokenization.md -------------------------------------------------------------------------------- /docs/chapter4/apply-bert.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/apply-bert.ipynb -------------------------------------------------------------------------------- /docs/chapter4/apply-gpt.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/apply-gpt.ipynb -------------------------------------------------------------------------------- /docs/chapter4/images/C4images1.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/images/C4images1.PNG -------------------------------------------------------------------------------- /docs/chapter4/images/C4images2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/images/C4images2.png -------------------------------------------------------------------------------- /docs/chapter4/images/C4images3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/images/C4images3.png -------------------------------------------------------------------------------- /docs/chapter4/images/Token1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/images/Token1.png -------------------------------------------------------------------------------- /docs/chapter4/images/Token2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/images/Token2.png -------------------------------------------------------------------------------- /docs/chapter4/task4.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter4/task4.md -------------------------------------------------------------------------------- /docs/chapter5/Transformer组件实现.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/Transformer组件实现.ipynb -------------------------------------------------------------------------------- /docs/chapter5/project.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/project.md -------------------------------------------------------------------------------- /docs/chapter5/task5.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/task5.md -------------------------------------------------------------------------------- /docs/chapter5/test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/test.py -------------------------------------------------------------------------------- /docs/chapter5/使用 NumPy 和 SciPy 实现通用注意力机制.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/使用 NumPy 和 SciPy 实现通用注意力机制.ipynb -------------------------------------------------------------------------------- /docs/chapter5/多头注意力机制.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/多头注意力机制.ipynb -------------------------------------------------------------------------------- /docs/chapter5/实践项目.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/实践项目.ipynb -------------------------------------------------------------------------------- /docs/chapter5/实践项目_GPU.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/实践项目_GPU.ipynb -------------------------------------------------------------------------------- /docs/chapter5/自注意力机制实现.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter5/自注意力机制实现.ipynb -------------------------------------------------------------------------------- /docs/chapter6/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter6/README.md -------------------------------------------------------------------------------- /docs/chapter6/Transformer_for_vision.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/fun-transformer/HEAD/docs/chapter6/Transformer_for_vision.md --------------------------------------------------------------------------------