DeepSpeed

58241b1d - fix: handle exception when loading cache file in test_inference.py (#5802)

fix: handle exception when loading cache file in test_inference.py (#5802) This PR is to fix CI failures such as https://github.com/microsoft/DeepSpeed/actions/runs/10085903860/job/27887546470#step:8:3616 cc @tjruwase Co-authored-by: Logan Adams <114770087+loadams@users.noreply.github.com>