Update README.md

Author: robmarkcole

README.md

@@ -1571,6 +1571,7 @@ For supervised machine learning, you will require annotated images. For example
 Also check the section **Image handling, manipulation & dataset creation**
 * [GroundWork](https://groundwork.azavea.com/) is designed for annotating and labeling geospatial data like satellite imagery, from Azavea
 * [labelbox.com](https://labelbox.com/) -> free tier is quite generous, supports annotating Geotiffs & returning annotations with geospatial coordinates. Watch [this webcast](https://www.arturo.ai/webcastbuilding-ai-products-from-the-ground-up/)
+* [diffgram](https://github.com/diffgram/diffgram) describes itself as a complete training data platform for machine learning delivered as a single application, supports [streaming data to pytorch & tensorflow](https://medium.com/diffgram/stream-training-data-to-your-models-with-diffgram-f0f25f6688c5). [COGS can be annotated](https://diffgram.readme.io/docs/geospatial-annotation-guide)
 * [iris](https://github.com/ESA-PhiLab/iris) -> Tool for manual image segmentation and classification of satellite imagery
 * If you are considering building an in house annotation platform [read this article](https://medium.com/earthcube-stories/ai-products-and-remote-sensing-yes-it-is-hard-and-yes-you-need-a-good-infra-4b5d6cf822f1). Used PostGis database, GeoJson format and GIS standard in a stateless architecture
 * [satellite-imagery-labeling-tool](https://github.com/microsoft/satellite-imagery-labeling-tool) -> from Microsoft, this is a lightweight web-interface for creating and sharing vector annotations over satellite/aerial imagery scenes
@@ -1593,7 +1594,6 @@ Also check the section **Image handling, manipulation & dataset creation**
 * SuperAnnotate can be run [locally](https://github.com/opencv-ai/superannotate) or used via a [cloud service](https://superannotate.com/)
 * [dash_doodler](https://github.com/dbuscombe-usgs/dash_doodler) -> A web application built with plotly/dash for image segmentation with minimal supervision
 * [remo](https://remo.ai) -> A webapp and Python library that lets you explore and control your image datasets
-* [diffgram](https://github.com/diffgram/diffgram) describes itself as a complete training data platform for machine learning delivered as a single application, supports [streaming data to pytorch & tensorflow](https://medium.com/diffgram/stream-training-data-to-your-models-with-diffgram-f0f25f6688c5)
 * TensorFlow Object Detection API provides a [handy utility](https://github.com/tensorflow/models/blob/6a55ecdea7afda51f9dc42dc17104bd6444395d9/research/object_detection/utils/colab_utils.py#L384) for object annotation within Google Colab notebooks. See usage [here](https://github.com/yasserius/tf2-object-detection-api#label-images-in-colab)
 * [coco-annotator](https://github.com/jsbroks/coco-annotator) -> Web-based image segmentation tool for object detection, localization, and keypoints
 * [pylabel](https://github.com/pylabel-project/pylabel) -> Python library for computer vision labeling tasks. The core functionality is to translate bounding box annotations between different formats-for example, from coco to yolo. PyLabel also includes an image labeling tool that runs in a Jupyter notebook that can annotate images manually or perform automatic labeling using a pre-trained model