--- title: 校准执行过程中提示"IfmrQuantCalibration with offset scale is illegal"或"IfmrQuantCalibration without offset sc description: 在调用Caffe框架执行中间校准模型推理过程中，由于输入数据范围不合法，导致量化算法计算得到的scale不合理，从而校准过程失败，终止Caffe校准流程。 keywords: IfmrQuantCalibration;illegal;without url: https://www.hiascend.com/document/caselibrary/detail/atlasamct_16_0376 section: (其他) --- # 校准执行过程中提示"IfmrQuantCalibration with offset scale is illegal"或"IfmrQuantCalibration without offset sc URL: https://www.hiascend.com/document/caselibrary/detail/atlasamct_16_0376 描述: 在调用Caffe框架执行中间校准模型推理过程中，由于输入数据范围不合法，导致量化算法计算得到的scale不合理，从而校准过程失败，终止Caffe校准流程。 关键词: IfmrQuantCalibration;illegal;without 昇腾文档 [了解详情](https://www.hiascend.com/document) 故障案例 [了解详情](https://www.hiascend.com/document/caselibrary) 校准执行过程中提示"IfmrQuantCalibration with offset scale is illegal"或"IfmrQuantCalibration without offset scale is illegal" 校准执行过程中提示"IfmrQuantCalibration with offset scale is illegal"或"IfmrQuantCalibration without offset scale is illegal" 2025/04/17 496 问题信息 | 问题来源 | 产品大类 | 产品子类 | 关键字 | | --- | --- | --- | ---| | 官方 | 应用开发 | CANN | IfmrQuantCalibration、illegal、without | #### 问题现象描述 在调用Caffe框架执行中间校准模型推理过程中，由于输入数据范围不合法，导致量化算法计算得到的scale不合理，从而校准过程失败，终止Caffe校准流程。 #### 原因分析 1. 数据范围 [-inf , +inf]： 因为AMCT的量化算法需要强制过零点，所以计算出的scale也就是inf/255=inf，该情况下量化因子后续无法承载，因此量化算法会提示错误信息，不支持该数据范围。 1. 场景1 ``` I0513 16:12:37.508546 80719 ifmr_layer.cu:58] Doing layer: "Convolution1" calibration, already store 1/1 data. I0513 16:12:37.508558 80719 ifmr_layer.cu:65] Start to do ifmr quant. [ERROR][ProcessScale][48] Not support scale is +inf. [ERROR][IfmrQuantCalibration][338] IfmrQuantCalibration with offset scale is illegal. [ERROR][IfmrQuantInternel][462] IfmrQuantInternel calibration failed. F0513 16:12:37.535028 80719 ifmr_layer.cu:70] Check failed: ret == 0 (-65519 vs. 0) Do IFMR calibration failed *** Check failure stack trace: *** Aborted ``` 2. 场景2 ``` I0513 16:14:48.173506 81053 ifmr_layer.cu:58] Doing layer: "Convolution1" calibration, already store 1/1 data. I0513 16:14:48.173519 81053 ifmr_layer.cu:65] Start to do ifmr quant. [ERROR][ProcessScale][48] Not support scale is +inf. [ERROR][IfmrQuantCalibration][359] IfmrQuantCalibration without offset scale is illegal. [ERROR][IfmrQuantInternel][462] IfmrQuantInternel calibration failed. F0513 16:14:48.199769 81053 ifmr_layer.cu:70] Check failed: ret == 0 (-65519 vs. 0) Do IFMR calibration failed *** Check failure stack trace: *** Aborted ``` 2. 数据范围：（其中EPSILON包括DBL_EPSILON double类型，FLT_EPSILON float类型，当前使用的是FLT_EPSILON类型） AMCT量化支持计算得到的最大，因为在昇腾AI处理器量化动作做的是乘法计算：， 如果scale大于，会小于FLT_EPSILON，此时量化后结果就不可信。因此AMCT量化算法仅支持原始数据范围在内进行量化，否则会提示不支持并提示错误信息。 ``` 2021-05-11 09:25:42,222 - INFO - [AMCT]:[WeightsCalibrationPass]: Do layer 'Convolution1[arq_quantize]' weights calibration success! 2021-05-11 09:25:42,222 - INFO - [AMCT]:[Optimizer]: Do 2021-05-11 09:25:42,223 - INFO - [AMCT]:[Optimizer]: Do 2021-05-11 09:25:42,223 - INFO - [AMCT]:[Optimizer]: Do 2021-05-11 09:25:42,223 - INFO - [AMCT]:[Graph]: Doing whole model dump... 2021-05-11 09:25:42,225 - INFO - [AMCT]:[Utils]: The weights_file is saved in xx/data/AMCT15_CAFFE_GPU_GPU_FAQ_LEVEL1_TC_001/AMCT15_CAFFE_GPU_GPU_FAQ_LEVEL1_TC_001_tmp.caffemodel 2021-05-11 09:25:42,226 - INFO - [AMCT]:[Utils]: The model_file is saved in xx/data/AMCT15_CAFFE_GPU_GPU_FAQ_LEVEL1_TC_001/AMCT15_CAFFE_GPU_GPU_FAQ_LEVEL1_TC_001_tmp.prototxt [ERROR][ProcessScale][52] Not support scale greater than 1 / FLT_EPSILON. [ERROR][IfmrQuantCalibration][338] IfmrQuantCalibration with offset scale is illegal. [ERROR][IfmrQuantInternel][462] IfmrQuantInternel calibration failed. WARNING: Logging before InitGoogleLogging() is written to STDERR F0511 09:25:42.292863 80432 ifmr_layer.cu:70] Check failed: ret == 0 (-65519 vs. 0) Do IFMR calibration failed *** Check failure stack trace: *** Aborted ``` #### 解决措施 根据提示信息，跳过日志中的量化层，例如上述提示信息中的Convolution1层。 本页内容 - 问题信息 - 问题现象描述 - 原因分析 - 解决措施