Closed Sudharsan-V closed 7 months ago
This commit contains benchmarking of SSD using coco128 dataset ( moved to weka path - /mnt/MLPerf/tt_dnn-models/ssd/coco128 )
Note: The model's low PCC is affecting the average precision during benchmarking.
As the inference time is exceeding 600 seconds with the complete dataset, only 50 samples are currently used.
Metrics used:
PR is merged, hence closing the ticket.
GS e2e Demo
test uses batched real inputs and produces batched outputs
this test is looped (takes several batches of inputs, produces several batches of outputs)
test evaluates results of TT model against validation data in selected data set
test is documented
how to run test (commands to execute) what does the model do (high-level explanation in a couple of sentences) description of what is input and what is output this demo should load inputs and expected outputs from a data set on Weka (TBD which one)