#embeddings #ai #cheatsheet Created at 130723 # [Anonymous feedback](https://www.admonymous.co/louis030195) # [[Epistemic status]] #shower-thought Last modified date: 130723 Commit: 0 # Related # ML embeddings cheatsheet Here's a cheatsheet of different methods and models to generate embeddings for various types of data, including text, image, and multimodal data: | Data Type | Model/Method | Description | |-----------|--------------|-------------| | Text | Word2Vec[5] | A shallow neural network model that generates word embeddings by learning from the context of words in large text corpora. | | Text | GloVe[6] | Global Vectors for Word Representation is an unsupervised learning algorithm that generates word embeddings by aggregating global word-word co-occurrence statistics from a corpus. | | Text | BERT[10] | Bidirectional Encoder Representations from Transformers is a pre-trained model that generates contextualized word embeddings using the transformer architecture. | | Image | VGG-16[4] | A pre-trained model for image classification that can be used to generate image embeddings by extracting features from the last fully connected layer. | | Image | ResNet50[4] | A pre-trained model for image classification with residual connections, which can be used to generate image embeddings by extracting features from the last fully connected layer. | | Image | Inceptionv3[4] | A pre-trained model for image classification with an inception module, which can be used to generate image embeddings by extracting features from the last fully connected layer. | | Multimodal | CLIP[3] | Contrastive Language-Image Pretraining is a model that learns to generate embeddings for both text and images in a shared semantic space, enabling tasks like zero-shot image classification and text-to-image search. | | Multimodal | ImageBind[13] | A model that generates multimodal word embeddings by combining text and visual information from large-scale web-annotated images. | These models and methods can be used to generate embeddings for various tasks, such as text classification, image classification, semantic search, and more. Note that some models may require fine-tuning or adaptation for specific tasks or domains. Citations: [1] http://cazabetremy.fr/Teaching/CN2021/CheatSheet/CN_CS_embedding.pdf [2] https://www.markdownguide.org/extended-syntax/ [3] https://stats.stackexchange.com/questions/319165/what-is-a-multimodal-embedding [4] https://medium.com/@junxie2/top-5-pre-trained-model-for-image-embedding-650755baeb2e [5] https://techblog.ezra.com/different-embedding-models-7874197dc410 [6] https://www.kdnuggets.com/2021/11/guide-word-embedding-techniques-nlp.html [7] https://www.tablesgenerator.com/markdown_tables [8] https://arxiv.org/abs/2110.02577 [9] https://towardsdatascience.com/top-10-pre-trained-models-for-image-embedding-every-data-scientist-should-know-88da0ef541cd [10] https://huggingface.co/blog/mteb [11] http://sleep.dashnine.org/common.html [12] https://www.w3schools.io/file/markdown-table/ [13] https://cbmm.mit.edu/sites/default/files/publications/6590-training-and-evaluating-multimodal-word-embeddings-with-large-scale-web-annotated-images.pdf [14] https://cloud.google.com/vertex-ai/docs/generative-ai/embeddings/get-image-embeddings [15] https://www.tensorflow.org/text/guide/word_embeddings [16] https://medium.com/nlplanet/two-minutes-nlp-sentence-transformers-cheat-sheet-2e9865083e7a [17] https://tabletomarkdown.com/format-markdown-table/ [18] https://weaviate.io/blog/multimodal-models [19] https://www.analyticsvidhya.com/blog/2022/07/recommending-similar-images-using-image-embedding/ [20] https://www.deepset.ai/blog/the-beginners-guide-to-text-embeddings [21] https://www.kaggle.com/code/rftexas/nlp-cheatsheet-master-nlp [22] https://www.codecademy.com/resources/docs/markdown/tables [23] https://encord.com/blog/imagebind-embedding-model-explained/ [24] https://developers.google.com/mediapipe/solutions/vision/image_embedder [25] https://machinelearningmastery.com/what-are-word-embeddings/ [26] https://github.com/adam-p/markdown-here/wiki/Markdown-Cheatsheet [27] https://www.pluralsight.com/guides/working-tables-github-markdown [28] https://sail.usc.edu/publications/files/Tseng-SPL2021.pdf [29] https://www.activeloop.ai/resources/generate-image-embeddings-using-a-pre-trained-cnn-and-store-them-in-hub/ [30] https://en.wikipedia.org/wiki/Word_embedding [31] https://zeroshotsgiven.substack.com/p/cheatsheet-for-learning-ai-embeddings [32] https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/organizing-information-with-tables [33] https://assets.amazon.science/89/bf/661d950d4059930c8f1d2e449ac6/joint-visual-textual-embedding-for-multimodal-style-search.pdf [34] https://www.pinecone.io/learn/image-search [35] https://openai.com/blog/new-and-improved-embedding-model [36] https://www.kdnuggets.com/2023/03/multimodal-models-explained.html [37] https://huggingface.co/blog/image-similarity [38] https://medium.com/nlplanet/two-minutes-nlp-11-word-embeddings-models-you-should-know-a0581763b9a9 [39] https://theaisummer.com/vision-language-models/