Towhee Examples are used to analyze the unstructured data with towhee, such as reverse image search, reverse video search, audio classification, question and answer systems, molecular search, etc.
Report Bug or Request Feature
x2vec, Towhee is all you need! Towhee can generate embedding vectors via a pipeline of ML models and other operations. It aims to make democratize x2vec, allowing everyone - from beginner developers to large organizations - to generate dense embeddings with just a few lines of code.
There are many interesting examples that use Towhee to process various unstructured data like images, audio, video, etc. You can easily run these examples on your machine.
| Bootcamp | Operators | |
| Getting Started | Getting Started with Towhee
An introduction to DataCollection, which can help you better learn the data processing pipeline with Towhee. |
|
| Image | Reverse Image Search Search for images that are similar or related to the input image, it supports a lot of models such as ResNet, VGG, EfficientNet, ViT, etc. |
Image Embedding Timm |
|
Image Animation Convert an image into an animated image. |
Animegan Cartoongan |
|
|
Image Deduplication Find exact or near-exact duplicates within a collection of images. |
Image Decode Timm |
|
|
Text Image Search Returns images related to the description of the input query text, which is cross-modal retrieval. |
CLIP |
|
|
Visualization Under the hood: Embedding models and ANNS indexes in image search. |
embedding_visualization |
|
| NLP | Q&A System Process user questions and give answers through natural language technology. |
Text Embedding DPR |
|
Recommendation System(TBD) Predict the "rating" or "preference" a user would give to an item. |
||
| Video | Reverse Video Search It takes a video as input to search for similar videos. |
Video Classification Pytorchvideo |
|
Video Classification
Video Classification is the task of producing a label that is relevant to the video given its frames. |
Video Classification |
|
|
Text Video Search Search for similar or related videos with the input text. |
CLIP4Clip |
|
|
Deepfake Detection Predict the probability of a fake video for a given video. |
Deepfake |
|
| Audio | Audio Classification Categorize certain sounds into certain categories, such as ambient sound classification and speech recognition. |
Audio Classification |
| Medical | Molecular Search
Search for similar molecular formulas based on the Tanimoto metric, and also supports searching for substructures and superstructures. |
RDKit |
| Data Science | Credit Card Approval Prediction
Predict whether the bank issues a credit card to the applicant, and the credit scores can objectively quantify the magnitude of risk. |
Logistic Regression Decision Tree SVM |
| Training | Fine Tune Tutorial about how to fine tuen with towhee. |
Image Embedding |
Contributions to Milvus Bootcamp are welcome from everyone. See Guidelines for Contributing for details.
Join the Towhee community on Slack to give feedback, ask for advice, and direct questions to the engineering team. You can also submit Issues or join Discussions.