From 0cd9ec75728fafbbbda11aee0178c83cb54eb5b7 Mon Sep 17 00:00:00 2001 From: Milvus-doc-bot Date: Mon, 13 Jan 2025 08:10:45 +0000 Subject: [PATCH] Release new docs to master --- .../site/en/integrations/integrate_with_pytorch.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/v2.5.x/site/en/integrations/integrate_with_pytorch.md b/v2.5.x/site/en/integrations/integrate_with_pytorch.md index 515b17fb1d..73813dd181 100644 --- a/v2.5.x/site/en/integrations/integrate_with_pytorch.md +++ b/v2.5.x/site/en/integrations/integrate_with_pytorch.md @@ -1,14 +1,14 @@ --- id: integrate_with_pytorch.md -summary: This page discusses image search using Milvus -title: Image Search with Milvus - Integration +summary: This page demostrates how to build image search with PyTorch and Milvus +title: Image Search with PyTorch and Milvus --- -# Image Search with Milvus +# Image Search with PyTorch and Milvus -On this page, we are going to be going over a simple image search example using Milvus. The dataset we are searching through is the Impressionist-Classifier Dataset found on [Kaggle](https://www.kaggle.com/datasets/delayedkarma/impressionist-classifier-data). For this example, we have rehosted the data in a public google drive. +This guide introduces an example of integrating PyTorch and Milvus to perform image search using embeddings. PyTorch is a powerful open-source deep learning framework widely used for building and deploying machine learning models. In this example, we’ll leverage its Torchvision library and a pre-trained ResNet50 model to generate feature vectors (embeddings) that represent image content. These embeddings will be stored in Milvus, a high-performance vector database, to enable efficient similarity search. The dataset used is the Impressionist-Classifier Dataset from [Kaggle](https://www.kaggle.com/datasets/delayedkarma/impressionist-classifier-data). By combining the deep learning capabilities of PyTorch with the scalable search functionality of Milvus, this example demonstrates how to build a robust and efficient image retrieval system. -For this example, we are just using the Torchvision pre-trained Resnet50 model for embeddings. Let's get started! +Let’s get started! ## Installing the requirements @@ -247,4 +247,4 @@ plt.savefig('search_result.png') The search result image should be similar to the following: -![Image search output](../../../assets/integrate_with_pytorch.png) \ No newline at end of file +![Image search output](../../../assets/integrate_with_pytorch.png)