ID 原文 译文
40596 将权重和偏置参数分别融合、最优化量化，减少了内存占用. The weight and bias parameters are separately fused, optimized and quantized to reduce memory usage.
40597 通过以VGG16作为测试网络，在识别Imagenet数据集时，精度仅损失了0.02%，在200 MHz的情况下，吞吐率达到了129.6 GOPS，功耗仅为5.26 W. By using VGG16 as the test network, when identifying the Imagenet data set, the accuracy was only lost by 0.02%. In the case of 200 MHz, the throughput rate reached 129.6 GOPS and the power consumption was only 5.2 6 W.
40598 提出了一种SATA固态硬盘Power Cycle功能自动化测试方法，用于验证SATA固态硬盘在上下电过程中数据完整性与安全性，保证SATA固态硬盘在多种应用场景中都可以保持正常且高效的运行状态. An automated test method for Power Cycle functionof SATA solid state drive(SSD)is proposed to verify the data integrity and safety of the SATA SSD in the processof power-on and power-off, and ensure that the SATA SSD can maintain normal and efficient operating state in various application scenarios.
40599 通过分析标准SATA协议与ATA命令集协议，针对SATA固态硬盘Power Cycle工作流程，在DriveMaster和ULink Power Hub软硬件平台基础上进行脚本编程，对SATA固态硬盘Power Cycle功能进行自动化测试. By analyzing the standard SATA protocol and focusing on the power cycle workflow of SATA SSD, the automatic test of SATA SSD Power Cycle function is carried out by scripting on the hardware and software platforms of DriveMaster and Ulink Power Hub.
40600 测试方法包含基于标准协议提炼出的测试流程图，软件平台DriveMaster脚本编程和DriveMaster平台与ULink Power Hub硬件组件联动设置，待测硬盘运行状况记录，发生异常时运行环境记录，测试结果汇总等内容. The test method includes the test flow chart extracted based on the SATA standard protocol, the software platform Drive Master script programming and the linkage setting of the Drive Master platform and the ULink Power Hub hardware components, the running status record of the device under test, the running environment record when an abnormality occurs, the test result summary, etc.
40601 为满足用户对SATA固态硬盘使用需求，测试过程在经典用例"写入数据-下电-上电-读回比较"基础上，为SATA固态硬盘制造多种不同写入压力，模拟SATA固态硬盘真实使用场景，从厂商角度保证SATA固态硬盘中数据完整性以及正确性. To meet the user demand for SATA SSD, the test process is based on the classic use case "write data-power off-power on-read back and comparison" to create a variety of different writing pressures for SATA SSD, simulating the real use of SATA SSD Scenarios to ensure the integrity and correctness of the data in the SATA SSD from the manufacturer's perspective.
40602 非均衡数据(分类不均匀分布)和冗余特征的出现极大增加了数据准确分类的难度. The unbalanced data(non-uniform distribution of classes) and the redundant features dramatically increased the difficulty of data accurate classification.
40603 以最优化学习算法的预测准确率为目标，结合合成少数过采样技术SMOTE，设计了一种针对非均衡数据的二进制排队搜索方法的包装式特征选择算法BQSA. Taking the prediction accuracy of the optimal learning algorithm as the goal, combined with the synthetic minority oversampling technology SMOTE, a wrapper feature selection algorithm BQSA was desigend for binary queue search method of unbalanced data.
40604 利用PROMISE知识库中十四种软件故障预测数据集进行实验分析.测试了数据集过采样比例的影响，证实合成少数过采样对高度非均衡数据的分类预测具有正面影响，并得到了最佳过采样率; Using 14 kinds of software fault prediction in PROMISE knowledge base to conduct experimental analysis of datasets. The influence of the over-sampling ratio of the dataset is tested, and it is proved that the synthesis of a few over-sampling has a positive effect on the classification prediction of highly unbalanced data, and the optimal over-sampling rate is obtained.
40605 比较了BQSA与同类算法的性能，证实结合合成少数过采样技术的BQSA算法拥有更优的预测准确性，在分类敏感度、专一性以及曲线下面积AUC等指标上表现更佳. The performance of BQSA is compared with similar algorithms, and it is proved that the BQSA algorithm combined with synthetic minority oversampling has better prediction accuracy and better performance in classification sensitivity, specificity and AUC of area under the curve.