Home
last modified time | relevance | path

Searched refs:bias (Results 1 – 25 of 47) sorted by relevance

12

/ohos5.0/docs/zh-cn/device-dev/reference/hdi-apis/nnrt/
H A D_full_connection_v10.md18 - bias,全连接的偏置,在量化场景下不需要量化参数,其量化版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决定。
36 | boolean [hasBias](#hasbias) | 是否使用bias。 |
78 是否使用bias
H A D_full_connection_v20.md18 - bias,全连接的偏置,在量化场景下不需要量化参数,其量化版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决定。
36 | boolean [hasBias](#hasbias) | 是否使用bias。 |
78 是否使用bias
H A D_conv2_d_fusion_v10.md18 - bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决…
H A D_conv2_d_fusion_v20.md18 - bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决…
H A D_conv2d_transpose_fusion_v10.md18 - bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决…
H A D_conv2d_transpose_fusion_v20.md18 - bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 int32 类型数据,实际量化参数由 x 和 weight 共同决…
H A D_bias_add_v10.md14 - bias,偏置值tensor。
H A D_bias_add_v20.md14 - bias,偏置值tensor。
H A D_scale_fusion_v10.md16 - bias,偏置tensor。
H A D_scale_fusion_v20.md16 - bias,偏置tensor。
/ohos5.0/docs/zh-cn/application-dev/reference/apis-arkui/arkui-ts/
H A Dts-container-relativecontainer.md18 …* 对齐后需要额外偏移可设置offset(API Version 11上新增了[bias](ts-universal-attributes-location.md#bias对象说明), 不建议再使…
26 …hainMode](ts-universal-attributes-location.md#chainmode12)接口;链内元素的bias属性全部失效,链头元素的bias作为整个链的bias生效。
27 …l-attributes-location.md#chainstyle12)链中,超出部分的分布可以通过[bias](ts-universal-attributes-location.md#bia…
417 本示例通过[bias](ts-universal-attributes-location.md#bias对象说明)实现了子组件的位置在竖直方向的两个锚点间偏移的效果
435 bias: { vertical: 0.3 }
707 …e](ts-universal-attributes-location.md#chainmode12)和[bias](ts-universal-attributes-location.md#bia…
727 bias: { horizontal: 0 }
H A Dts-universal-attributes-location.md147 | bias<sup>11+</sup> | [Bias](#bias对象说明) | 设置组件在锚点约束下的偏移参数,其值为到左/上侧锚点的距离与锚点间总距离的比值。<br/>**卡片能力:**…
161 | bias | [Bias](#bias对象说明) | 设置组件在锚点约束下的偏移参数,其值为到左/上侧锚点的距离与锚点间总距离的比值。|
194 | horizontal | number | &nbsp;否 | 水平方向上的bias值。<br/>当子组件的width属性有正确值并且有2个水平方向的锚点时生效。<br/>默认值:&nbsp;…
195 | vertical | number | &nbsp;否 | 垂直方向上的bias值。<br/>当子组件的height属性有正确值并且有2个垂直方向的锚点时生效。<br/>默认值:&nbsp;0…
/ohos5.0/foundation/graphic/graphic_3d/lume/Lume_3D/api/3d/shaders/common/
H A D3d_dm_shadowing_common.h69 const float bias = max(normalBias * (1.0 - NoL), depthBias); in CalcPcfShadow() local
89 const float compareDepth = shadowCoord.z - bias; in CalcPcfShadow()
141 const float bias = max(normalBias * (1.0 - NoL), depthBias); in CalcPcfShadowMed() local
161 const float compareDepth = shadowCoord.z - bias; in CalcPcfShadowMed()
200 const float bias = 0.002; in CalcPcfShadowSimpleSample() local
202 const float compareDepth = shadowCoord.z - bias; in CalcPcfShadowSimpleSample()
/ohos5.0/docs/zh-cn/release-notes/changelogs/OpenHarmony_4.1.5.1/
H A Dchangelogs-arkui.md153 当同时设置了组件尺寸和约束条件时,优先使用开发者设置的尺寸来表示组件大小。同时在API version 11上,增加了bias属性用于设置组件在约束条件下的偏移,上述约定也更便于开发者理解bias的…
/ohos5.0/docs/en/application-dev/reference/apis-arkui/arkui-ts/
H A Dts-container-relativecontainer.md18 …required after the alignment, it can be set through [bias](ts-universal-attributes-location.md#bia…
26 …onent. The **bias** attributes of all elements in the chain are ineffective, and the bias of the c…
27 … distribution of the excess part can be set through [bias](ts-universal-attributes-location.md#bia…
417 This example uses the [bias](ts-universal-attributes-location.md#bias) API to create a vertical off…
435 bias: { vertical: 0.3 }
707 …s-universal-attributes-location.md#chainmode12) and [bias](ts-universal-attributes-location.md#bia…
727 bias: { horizontal: 0 }
/ohos5.0/foundation/arkui/ace_engine/frameworks/core/components_ng/pattern/relative_container/
H A Drelative_container_layout_algorithm.cpp403 BiasPair bias(0.5f, 0.5f); in CheckHorizontalChain() local
406 bias = flexItem->GetBiasValue(); in CheckHorizontalChain()
433 chainParam.bias = bias; in CheckHorizontalChain()
491 BiasPair bias(0.5f, 0.5f); in CheckVerticalChain() local
496 bias = flexItem->GetBiasValue(); in CheckVerticalChain()
523 chainParam.bias = bias; in CheckVerticalChain()
694 …float bias = (direction == LineDirection::HORIZONTAL) ? chainParam.bias.first : chainParam.bias.se… in CalcOffsetInChainGetStart() local
707 start = (anchorDistance - contentSize) * bias; in CalcOffsetInChainGetStart()
719 start = (anchorDistance - contentSize) * bias; in CalcOffsetInChainGetStart()
1458 bool RelativeContainerLayoutAlgorithm::IsValidBias(float bias) in IsValidBias() argument
[all …]
H A Drelative_container_layout_algorithm.h67 BiasPair bias; member
130 bool IsValidBias(float bias);
/ohos5.0/docs/zh-cn/application-dev/reference/apis-neural-network-runtime-kit/
H A D_neural_network_runtime.md580 | OH_NN_OPS_BIAS_ADD | 对给出的输入张量上的各个维度方向上的数据进行偏置。<br/>输入:<br/>- input,输入张量,可为2-5维度。<br/>- bias,参数对应输…
583bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32类型数据,实际量化参数由input和weight…
584bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32 类型数据,实际量化参数由input和weig…
585bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32类型数据,实际量化参数由input和weight…
590bias,全连接的偏置,在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入OH_NN_INT32类型数据,实际量化参数由input和weight共同决定。<br/>参数:<br/>-…
601 | OH_NN_OPS_SCALE | 给定一个张量,计算其缩放后的值。<br/>输入:<br/>- input,一个n维张量。<br/>- scale,缩放张量。<br/>- bias,偏置张量。…
642 …,方差为1。<br/>输入:<br/>- input,4维张量。<br/>- scale,1维张量,缩放系数,尺寸和输入的通道数一致。<br/>- bias,1维张量,偏置常量,尺寸和输入的通道数…
650 …层到隐藏层的权重参数,形状为[numDirections\*numLayers, 4\*hiddenSize, inputSize]。<br/>- bias,输入和隐藏层到隐藏层的偏置,形状为[n…
671 …<br/>- input,4维张量,待归一化的输入张量。<br/>参数:<br/>- depthRadius,标量,归一化窗口的半宽。<br/>- bias,偏移量,通常为正避免除零问题,默认为1…
920 | OH_NN_LRN_BIAS<sup>12+</sup> | 当张量作为LRN算子的bias参数时,使用本枚举值。 |
H A D_neural_nework_runtime.md580 | OH_NN_OPS_BIAS_ADD | 对给出的输入张量上的各个维度方向上的数据进行偏置。<br/>输入:<br/>- input,输入张量,可为2-5维度。<br/>- bias,参数对应输…
583bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32类型数据,实际量化参数由input和weight…
584bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32 类型数据,实际量化参数由input和weig…
585bias,卷积的偏置,是长度为[outChannel]的数组。在量化场景下,bias参数不需要量化参数,其量化 版本要求输入 OH_NN_INT32类型数据,实际量化参数由input和weight…
590bias,全连接的偏置,在量化场景下,bias 参数不需要量化参数,其量化 版本要求输入OH_NN_INT32类型数据,实际量化参数由input和weight共同决定。<br/>参数:<br/>-…
601 | OH_NN_OPS_SCALE | 给定一个张量,计算其缩放后的值。<br/>输入:<br/>- input,一个n维张量。<br/>- scale,缩放张量。<br/>- bias,偏置张量。…
642 …,方差为1。<br/>输入:<br/>- input,4维张量。<br/>- scale,1维张量,缩放系数,尺寸和输入的通道数一致。<br/>- bias,1维张量,偏置常量,尺寸和输入的通道数…
650 …层到隐藏层的权重参数,形状为[numDirections\*numLayers, 4\*hiddenSize, inputSize]。<br/>- bias,输入和隐藏层到隐藏层的偏置,形状为[n…
671 …<br/>- input,4维张量,待归一化的输入张量。<br/>参数:<br/>- depthRadius,标量,归一化窗口的半宽。<br/>- bias,偏移量,通常为正避免除零问题,默认为1…
920 | OH_NN_LRN_BIAS<sup>12+</sup> | 当张量作为LRN算子的bias参数时,使用本枚举值。 |
/ohos5.0/foundation/filemanagement/storage_service/services/storage_daemon/file_sharing/test/
H A Dacl_test.cpp322 size_t bias = 1; variable
323 rc = acl2.DeSerialize(buf, sizeof(AclXattrHeader) + sizeof(AclXattrEntry) * 100 + bias);
/ohos5.0/drivers/interface/location/gnss/v2_0/
H A DGnssTypes.idl556 * Satellite clock bias value in meters.
561 * Satellite clock drift bias value in meters.
602 * Sub-nanosecond bias of receiverClockTime.
607 …* Uncertainty of the local estimate of GPS time (clock bias 'rcvClockFullBias' and 'rcvClockSubBia…
630 … * the clock bias ('rcvClockFullBias' + 'rcvClockSubBias' should be less than clockJumpThreshold)
/ohos5.0/docs/en/application-dev/device/sensor/
H A Dsensor-overview.md11 …X, Y, and Z), in the unit of m/s<sup>2</sup>.| Measuring the acceleration bias estimation …
15 …n three physical axes (X, Y, and Z), in the unit of rad/s.| Measuring the bias estimation of the r…
21 …ical axes (X, Y, and Z), in the unit of μT.| Measuring the magnetic field bias estimation …
/ohos5.0/foundation/window/window_manager/dmserver/src/
H A Dscreen_rotation_controller.cpp201 int32_t bias = static_cast<int32_t>(defaultDeviceRotationOffset_ / 90); in CalcDeviceRotation() local
202 int32_t deviceRotationValue = static_cast<int32_t>(sensorRotation) - bias; in CalcDeviceRotation()
/ohos5.0/docs/zh-cn/application-dev/reference/apis-sensor-service-kit/
H A Djs-apis-sensor.md107 console.info('Succeeded in invoking on. X-coordinate bias: ' + data.biasX);
108 console.info('Succeeded in invoking on. Y-coordinate bias: ' + data.biasY);
109 console.info('Succeeded in invoking on. Z-coordinate bias: ' + data.biasZ);
389 console.info('Succeeded in invoking on. X-coordinate bias: ' + data.biasX);
390 console.info('Succeeded in invoking on. Y-coordinate bias: ' + data.biasY);
391 console.info('Succeeded in invoking on. Z-coordinate bias: ' + data.biasZ);
671 console.info('Succeeded in invoking on. X-coordinate bias: ' + data.biasX);
672 console.info('Succeeded in invoking on. Y-coordinate bias: ' + data.biasY);
673 console.info('Succeeded in invoking on. Z-coordinate bias: ' + data.biasZ);
4770 console.info('Succeeded in invoking on. X-coordinate bias: ' + data.biasX);
[all …]
/ohos5.0/drivers/interface/nnrt/v2_1/
H A DNodeAttrTypes.idl300 * * <b>bias</b>, the bias tensor.
436 * * <b>bias</b>: bias of the convolution. It is an array with a length of <b>[outChannel]</b>.\n
437 * In the quantization scenario, <b>bias</b> does not need to be quantized.
499 * * <b>bias</b>: bias of the convolution. It is an array with a length of <b>[outChannel]</b>.\n
900 …* * <b>bias</b>, a full-connection bias. In quantization scenarios, a quantized parameter is not r…
913 /** Whether to use the bias. */
1033 * * <bias> The input 1-dimensional bias tensor of size C.
1252 float bias;
1299 * * <b>bias</b>, bias tensor of input-layer and hidden-layer to hidden-layer,
1320 /** Whether the operation contains bias. */
[all …]

12