Daniel Gleaves
Daniel Gleaves
For CIFAR-10-4K, you have the model train for 300,000 steps, while the original paper trains for 1,000,000 steps. What is the reason for stopping earlier? 
I come bringing an interesting issue. I am seeing issues similar to #215 in case anything has changed since then. Looking back at the logs, for the location that uses...
https://github.com/temporalio/docker-builds/issues/222 The 1.24 tag for temporal admin tools does not work in the quick start guide's docker-compose file. The latest is temporalio/admin-tools:1.24.2-tctl-1.18.1-cli-1.0.0 -- there is no 1.24 tag.
### Search before asking - [x] I have searched the Inference [issues](https://github.com/roboflow/inference/issues) and found no similar bug report. ### Bug https://github.com/roboflow/inference/blob/295ca683ff029e6482cc859811514d4c7d1e2d59/inference/core/interfaces/stream/inference_pipeline.py#L484 Because the default profiling directory is found in "./inference_profiling"...