Abstract: A method of performing automatic tuning on a deep learning model includes: utilizing an instruction-based learned cost model to estimate a first type of operational performance metrics based on a tuned configuration of layer fusion and tensor tiling; utilizing statistical data gathered during a compilation process of the deep learning model to determine a second type of operational performance metrics based on the tuned configuration of layer fusion and tensor tiling; performing an auto-tuning process to obtain a plurality of optimal configurations based on the first type of operational performance metrics and the second type of operational performance metrics; and configure the deep learning model according to one of the plurality of optimal configurations.

Abstract: A barcode recognition method is provided for recognizing a barcode to derive effective barcode information thereof. The barcode includes a plurality of segments. In the method, every time the barcode is scanned and recognized, any segment, in which a barcode information fragment carried therein is able to be derived, is recorded as an effective segment. And then the effective segments are selected and combined to generate a plurality of segment combinations. The information carried by the segment combinations is respectively verified by the checksum thereof, so as to find out the segment combination, which is successfully verified by the checksum thereof, to be the effective combination. Every time the barcode is scanned and recognized, only part of the segments of the barcode has to be successfully recognized, and the effective segments derived are combined and verified, so as to quickly recognize the barcode.

Abstract: A radio frequency identification (RFID) tag tracking system is provided for performing an RFID tag tracking method. The RFID tag tracking system includes an RFID tag, at least one RFID reader, and a backend-server. The RFID tag periodically emits tag signals having different original signal strengths, and the original signal strength of each tag signal determines a signal receiving boundary relative to the RFID tag. The RFID reader is used to receive the tag signals, and the tag signal that can be received by the RFID reader receives the tag signals, and determines one or more receivable signals which cab be received by the RFID reader among the tag signals. The backend-server determines a possible position area of the RFID tag relative to the RFID reader according to the signal receiving boundary of the receivable signal.

Abstract: A barcode recognition method is provided for recognizing a barcode to derive effective barcode information thereof. The barcode includes a plurality of segments. In the method, every time the barcode is scanned and recognized, any segment, in which a barcode information fragment carried therein is able to be derived, is recorded as an effective segment. And then the effective segments are selected and combined to generate a plurality of segment combinations. The information carried by the segment combinations is respectively verified by the checksum thereof, so as to find out the segment combination, which is successfully verified by the checksum thereof, to be the effective combination. Every time the barcode is scanned and recognized, only part of the segments of the barcode has to be successfully recognized, and the effective segments derived are combined and verified, so as to quickly recognize the barcode.