Thus, if your model can be exported to ONNX or TFLite, it is almost certainly supported by Samtool, provided the operators are not too exotic.
SamsTool provides a suite of service functions for these models: samtool supported models
Leverages masked image pretraining (SAMI) for lightweight, fast inference. Thus, if your model can be exported to
# Sort using a memory buffer appropriate for the model's output samtools sort -@ 8 -m 4G -o sorted.bam output.bam samtool supported models
: Integrated functions that allow for the direct testing of management rules based on model outputs.
To illustrate the practical impact of Samtool’s optimization, here are inference latency numbers (ms per batch=1) on a Qualcomm Snapdragon 8 Gen 2 (Hexagon DSP) for key supported models: