Xing Yun (邢云)

Results 4 comments of Xing Yun (邢云)

The id is in '6-digit' format if you download stanford-cars via this link (https://aistudio.baidu.com/aistudio/datasetdetail/85765) instead of kaggle.

> 4 pytorch allocator cache flushes since last step. this happens when there is high memory pressure and is detrimental to performance. For eliminating this warning, I install `deepspeed` from...

For `deepspeed-0.12.6`, it can be specified in `deepspeed/runtime/zero/stage3.py`, in function `_optimizer_step` of `class DeepSpeedZeroOptimizer_Stage3`. Hope this may help you.

Perhaps you can try comment line 86 of third_party/Mask2Former/setup.py and install detectron2 manually. This works for me.