ResNet50_IN1k is an image embedding model designed for visual grouping and retrieval tasks. It is based on the ResNet-50 architecture and is intended to provide a dependable visual baseline for projects that require general image grouping. The model is suitable for use cases where reliable image embeddings are needed, particularly in contexts where projects benefit from established visual models.
The model accepts images as input and generates embeddings, specifically outputting a float32 tensor with a shape of [batch, 2048]. This output format is intended to support downstream tasks such as image retrieval and grouping. ResNet50_IN1k is categorized as a medium compute tier model, balancing quality and domain coverage with moderate runtime requirements. It is designed to run efficiently on recent laptops and desktops, making it accessible for a range of users without the need for high-end hardware.
ResNet50_IN1k is delivered in ONNX format and is compatible with onnxruntime, facilitating integration into various workflows and platforms that support ONNX models. The model is part of the eidora-model-zoo and is associated with the EIDORA platform. It is not recommended for use with text, video, or audio inputs, nor for fine-grained semantic retrieval tasks where self-supervised models might be preferred.
The model is released under the BSD 3-Clause license, allowing for broad use and distribution within the terms of this open-source license. 03385 for its underlying architecture. ResNet50_IN1k is positioned as a reliable tool for those seeking established visual baselines in image-related machine learning projects.
In the Foundation models & chat space, ResNet50 IN1k takes a focused approach. It provides a robust ResNet-50 image embedding model for visual grouping and retrieval tasks. ResNet50 IN1k is an open-source project aimed at computer vision researchers and developers. The project is open source (BSD-3-Clause). It runs on the web and the command line, and it can be self-hosted.
EIDORA builds and maintains ResNet50 IN1k, and the product first shipped in 2016. The project is developed in the open on GitHub with 17.8k stars and 30 commits in the last 90 days. Across PulseGate's embedding index, ResNet50 IN1k has few near neighbours, marking it as relatively distinct. Among its 5 catalogued features are image embedding, resNet-50 architecture, and ONNX support.
Latest indexed changes and source events
EIDORA/ResNet50_IN1k discovered by the PulseGate indexer
Other apps tracked under the same category.
