There are many image datasets available for training image classifiers, but here are a few that might be useful for your specific task of classifying indoor and outdoor scenes:
Places365: This dataset contains over 1.8 million images categorized into 365 scene categories, including indoor and outdoor scenes. You can download the dataset from the official website http://places2.csail.mit.edu/download.html
SUN Database: This dataset contains over 130,000 images categorized into 908 scene categories, including indoor and outdoor scenes. You can download the dataset from the official website http://vision.princeton.edu/projects/2010/SUN/
MIT Indoor Scenes: This dataset contains over 15,000 images categorized into 67 indoor scene categories. You can download the dataset from the official website http://web.mit.edu/torralba/www/indoor.html
Outdoor Scenes Dataset: This dataset contains over 14,000 outdoor images, which are classified into 8 different outdoor scene categories. You can download the dataset from the official website http://people.csail.mit.edu/celiu/OOD/OOD.html
ADE20K Dataset: This dataset contains over 20,000 images categorized into 150 scene categories, including indoor and outdoor scenes. You can download the dataset from the official website http://groups.csail.mit.edu/vision/datasets/ADE20K/
You can use any of these datasets to train an image classifier to classify indoor and outdoor scenes. You may need to preprocess and format the data to suit your specific requirements.