From d347f9bf22b449b1f6dcc6467f6bad265d0b9d67 Mon Sep 17 00:00:00 2001
From: xieting <xieting21@huawei.com>
Date: Tue, 5 Dec 2023 11:46:31 +0800
Subject: [PATCH] =?UTF-8?q?=E6=B7=BB=E5=8A=A0Tensor=20=E9=AD=94=E6=B3=95?=
 =?UTF-8?q?=E5=87=BD=E6=95=B0=E5=85=B7=E4=BD=93=E5=AF=B9=E5=BA=94=E6=93=8D?=
 =?UTF-8?q?=E4=BD=9CFAQ?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .../api_accuracy_checker/README.md            | 37 +++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/debug/accuracy_tools/api_accuracy_checker/README.md b/debug/accuracy_tools/api_accuracy_checker/README.md
index c2db434f71..36027998c5 100644
--- a/debug/accuracy_tools/api_accuracy_checker/README.md
+++ b/debug/accuracy_tools/api_accuracy_checker/README.md
@@ -172,3 +172,40 @@ API预检通过测试，则在accuracy_checking_details.csv文件中的“pass
 3. 为什么浮点型数据bench和npu的dtype不一致？
 
    答：对于fp16的数据，cpu会上升一个精度fp32去计算，这是和算子那边对齐的精度结论，cpu用更高精度去计算会更接近真实值
+
+4. Tensor 魔法函数具体对应什么操作？
+
+   答：
+
+   | Tensor魔法函数 | 具体操作 |
+   | --- | --- |
+   | `__add__` | + |
+   | `__and__` | & |
+   | `__bool__` | 返回Tensor布尔值 |
+   | `__div__` | / |
+   | `__eq__` | == |
+   | `__ge__` | >= |
+   | `__gt__` | > |
+   | `__iadd__` | += |
+   | `__iand__` | &= |
+   | `__idiv__` | /= |
+   | `__ifloordiv__` | //= |
+   | `__ilshift__` | <<= |
+   | `__imod__` | %= |
+   | `__imul__` | *= |
+   | `__ior__` | \|= |
+   | `__irshift__` | >>= |
+   | `__isub__` | -= |
+   | `__ixor__` | ^= |
+   | `__lshift__` | << |
+   | `__matmul__` | 矩阵乘法 |
+   | `__mod__` | % |
+   | `__mul__` | * |
+   | `__nonzero__` | 同`__bool__` |
+   | `__or__` | \| |
+   | `__radd__` | +（反向） |
+   | `__rmul__` | *（反向） |
+   | `__rshift__` | >> |
+   | `__sub__` | - |
+   | `__truediv__` | 同`__div__` |
+   | `__xor__` | ^ |
-- 
Gitee