inference icon indicating copy to clipboard operation
inference copied to clipboard

Reference implementations of MLPerf™ inference benchmarks

Results 200 inference issues
Sort by recently updated
recently updated
newest added

Change model name to a more explicit name "Stable Diffusion XL 1.0"

Issue Description: I manually download the model.pytorch and vocab.txt to the designated folder and run `/home/user/cm/bin/python3 run.py --backend=pytorch --scenario=Offline --max_examples 10 --mlperf_conf '/home/user/CM/repos/local/cache/b737554800c84148/inference/mlperf.conf' --user_conf '/home/user/CM/repos/mlcommons@ck/cm-mlops/script/generate-mlperf-inference-user-conf/tmp/c8eb2a31fd70402a93daee688bf391fa.conf' --accuracy 2>&1 | tee /home/user/CM/repos/local/cache/454869b45fbf4f67/test_results/default-reference-gpu-pytorch-v2.2.1-default_config/bert-99/offline/accuracy/console.out`...

`Loadgen built with uncommitted changes` error should be thrown only when loadgen-related files are modified. Currently this is thrown when any of the files in the inference repository is modified.

Hi there, just saw the video from the ai engineer Inference looks really awesome I've got this mini project in mind for years now, Getting a model to tell how...

I believe there is no need to run LLAMA2-70B model for 24576 samples. GPT-J-6B is run for 13368 samples, Stable diffusion is run for 5000 samples, 3d-unet is run for...

LlaMa2-70b has a run_accuracy.py script that as a last step that calls another script, consolidate_results.py. That last step does not work, it expects to read output pkl files, but those...

Currently the loadgen over the network is connecting only to the same host. This fix makes it connect to external SUTs in the network.