Docker Image Export/Import Instructions
1. Save Docker Image to File
After the build completes, save the image as a .tar file:
cd /data/usr/ft42/CVIT_XAI/LungRADS_Modeling/CLARITY/PiNS
docker save medical-imaging/nodule-segmentation:latest -o medical-imaging-nodule-segmentation.tar
2. Compress the Image (Optional - Recommended)
Compress the .tar file to save space:
gzip medical-imaging-nodule-segmentation.tar
# This creates: medical-imaging-nodule-segmentation.tar.gz
3. Check File Size
ls -lh medical-imaging-nodule-segmentation.tar*
4. Load Image on Another Machine
To use the saved image on another machine:
# If compressed:
gunzip medical-imaging-nodule-segmentation.tar.gz
# Load the image:
docker load -i medical-imaging-nodule-segmentation.tar
# Verify it loaded:
docker images | grep medical-imaging
5. Share the Image
You can now share the .tar.gz file:
- Copy to USB drive
- Upload to cloud storage
- Transfer via network
- Include in project distribution
File Size Expectations
- Uncompressed .tar: ~1.5-2.0 GB
- Compressed .tar.gz: ~800MB-1.2GB
Alternative: Docker Hub (Optional)
You can also push to Docker Hub for easy sharing:
# Tag the image
docker tag medical-imaging/nodule-segmentation:latest ft42/nodule-segmentation:latest
# Push to Docker Hub
docker push ft42/nodule-segmentation:latest
Usage After Import
Once loaded, use exactly as before:
./scripts/DLCS24_KNN_2mm_Extend_Seg.sh