nn.Embedding,IndexSelect,save,load:cpu,cuda 验证 (#29)

* benchmark.IR:支持对算子进行benchmark

* gather,save,load:cuda 验证完成

* IndexSelect:测试验证
RenameTensor：测试验证

front:embedding ok

* cuda:load,save,indexselect，和pytorch对齐

* nn.module:移除自动注册，改为手动注册tensor、module

Author: miaobyte

doc/excuter/op-mem-cuda/list.md

@@ -13,10 +13,11 @@
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| copytensor |  none  | copytensor(tensor<any> src, tensor<any> dst)->() | T2.data = T1.data | copytensor(tensor<any> src, tensor<any> dst)->() |
+| renametensor |  none  | renametensor(tensor<any> t, var<string> new_name)->() | rename T1 to T2 | renametensor(tensor<any> t, var<string> new_name)->() |
 | newtensor |  none  | newtensor(vector<int32> shape)->(tensor<any> tensor1) | T1 = zeros(shape) | newtensor(vector<int32> shape)->(tensor<any> tensor1) |
 | newtensor |  none  | newtensor(var<string> shape)->(tensor<any> tensor1) | T1 = zeros(shape) | newtensor(var<string> shape)->(tensor<any> tensor1) |
 | deltensor |  none  | deltensor(tensor<any> t)->() | del T1 | deltensor(tensor<any> t)->() |
+| copytensor |  none  | copytensor(tensor<any> src, tensor<any> dst)->() | T2.data = T1.data | copytensor(tensor<any> src, tensor<any> dst)->() |
 
 ### io
 
@@ -31,9 +32,9 @@
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| arange | miaobyte | arange(tensor<any> t, var<any> start, var<any> step)->() | arange(T1,start,step) | arange(tensor<any> t, var<any> start, var<any> step)->() |
 | normal | miaobyte | normal(tensor<any> t, var<any> mean, var<any> stddev, var<int32> seed)->() | normal(T1,mean,stddev,seed) | normal(tensor<any> t, var<any> mean, var<any> stddev, var<int32> seed)->() |
 | uniform | miaobyte | uniform(tensor<any> t, var<any> low, var<any> high, var<int32> seed)->() | uniform(T1,low,high,seed) | uniform(tensor<any> t, var<any> low, var<any> high, var<int32> seed)->() |
+| arange | miaobyte | arange(tensor<any> t, var<any> start, var<any> step)->() | arange(T1,start,step) | arange(tensor<any> t, var<any> start, var<any> step)->() |
 | constant | miaobyte | constant(tensor<any> t, var<any> value)->() | constant(T1) | constant(tensor<any> t, var<any> value)->() |
 
 ### elementwise
@@ -46,10 +47,10 @@
 | min | miaobyte | min(tensor<any> A, tensor<any> B)->(tensor<any> C) | T3=min(T1, T2) | min(tensor<any> A, tensor<any> B)->(tensor<any> C) |
 | maxscalar | miaobyte | maxscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=max(T1, scalar) | maxscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | tan | miaobyte | tan(tensor<float64|float32> A)->(tensor<float64|float32> C) | T3=tan(T1) | tan(tensor<float64|float32> A)->(tensor<float64|float32> C) |
-| divscalar | miaobyte | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=scalar/T1 | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | sin | miaobyte | sin(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) | T3=sin(T1) | sin(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) |
-| add | cublas | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
-| add | miaobyte | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
+| divscalar | miaobyte | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=scalar/T1 | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
+| log | miaobyte | log(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) | T3=log(T1) | log(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) |
+| addscalar | miaobyte | addscalar(tensor<any> A, var<any> b)->(tensor<any> C) | T3=T1+scalar | addscalar(tensor<any> A, var<any> b)->(tensor<any> C) |
 | greater | miaobyte | greater(tensor<any> A, tensor<any> B)->(tensor<bool> mask) | mask=compare(T1, T2) | greater(tensor<any> A, tensor<any> B)->(tensor<bool> mask) |
 | lessscalar | miaobyte | lessscalar(tensor<any> A, var<any> scalar)->(tensor<bool> mask) | mask=compare(T1, scalar) | lessscalar(tensor<any> A, var<any> scalar)->(tensor<bool> mask) |
 | cos | miaobyte | cos(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) | T3=cos(T1) | cos(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) |
@@ -58,8 +59,8 @@
 | minscalar | miaobyte | minscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=min(T1, scalar) | minscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | rdivscalar | miaobyte | rdivscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) | T3=scalar/T1 | rdivscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) |
 | rpowscalar | miaobyte | rpowscalar(var<float64|int32> scalar, tensor<float64|float32> A)->(tensor<float64|float32> C) | T3=pow(scalar, T1) | rpowscalar(var<float64|int32> scalar, tensor<float64|float32> A)->(tensor<float64|float32> C) |
-| log | miaobyte | log(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) | T3=log(T1) | log(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) |
-| addscalar | miaobyte | addscalar(tensor<any> A, var<any> b)->(tensor<any> C) | T3=T1+scalar | addscalar(tensor<any> A, var<any> b)->(tensor<any> C) |
+| add | cublas | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
+| add | miaobyte | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
 | sub | miaobyte | sub(tensor<any> A, tensor<any> B)->(tensor<any> C) | T3=T1-T2 | sub(tensor<any> A, tensor<any> B)->(tensor<any> C) |
 | sqrt | miaobyte | sqrt(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) | T3=sqrt(T1) | sqrt(tensor<float64|float32|float16|bfloat16> A)->(tensor<float64|float32|float16|bfloat16> C) |
 | subscalar | miaobyte | subscalar(tensor<any> A, var<any> b)->(tensor<any> C) | T3=T1-scalar | subscalar(tensor<any> A, var<any> b)->(tensor<any> C) |
@@ -78,22 +79,22 @@
 |-----------|--------|------------|--------------|----------------|
 | matmul | cublas | matmul(tensor<any> A, tensor<any> B)->(tensor<any> C) | T3=T1 @ T2 | matmul(tensor<any> A, tensor<any> B)->(tensor<any> C) |
 
+### reduce
+
+| Operation | Author | Func Def | Math Formula | IR Instruction |
+|-----------|--------|------------|--------------|----------------|
+| prod | miaobyte | prod(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = prod(A, axis=[1 2], keepdims=false) | prod(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
+| reducemax | miaobyte | reducemax(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = reducemax(A, axis=[1 2], keepdims=false) | reducemax(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
+| sum | miaobyte | sum(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = sum(A, axis=[1 2], keepdims=false) | sum(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
+| reducemin | miaobyte | reducemin(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = reducemin(A, axis=[1 2], keepdims=false) | reducemin(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
+
 ### changeshape
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| gather | miaobyte | gather(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) | T2 = T1.gather(indices=[1,2], axis=1) | gather(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) |
+| indexselect | miaobyte | indexselect(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) | T2 = T1.indexselect(index=[1,2], axis=1) | indexselect(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) |
 | broadcastTo | miaobyte | broadcastTo(tensor<any> A, vector<int32> new_shape)->(tensor<any> B) | T2 = T1.broadcastTo(new_shape=[4,3,2]) | broadcastTo(tensor<any> A, vector<int32> new_shape)->(tensor<any> B) |
 | concat | miaobyte | concat(listtensor<any> tensors, var<int32> dim)->(tensor<any> result) | Tresult = concat([T1, T2...], axis=3) | concat(listtensor<any> tensors, var<int32> dim)->(tensor<any> result) |
 | transpose | miaobyte | transpose(tensor<any> A, vector<int32> dim_order)->(tensor<any> C) | T2 = T1.transpose(dimorder=[1,0]) | transpose(tensor<any> A, vector<int32> dim_order)->(tensor<any> C) |
 | reshape | miaobyte | reshape(tensor<any> A, vector<int32> shape)->(tensor<any> B) | T1.reshape(shape)->T2 | reshape(tensor<any> A, vector<int32> shape)->(tensor<any> B) |
 
-### reduce
-
-| Operation | Author | Func Def | Math Formula | IR Instruction |
-|-----------|--------|------------|--------------|----------------|
-| reducemax | miaobyte | reducemax(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = reducemax(A, axis=[1 2], keepdims=false) | reducemax(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
-| prod | miaobyte | prod(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = prod(A, axis=[1 2], keepdims=false) | prod(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
-| sum | miaobyte | sum(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = sum(A, axis=[1 2], keepdims=false) | sum(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
-| reducemin | miaobyte | reducemin(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) | B = reducemin(A, axis=[1 2], keepdims=false) | reducemin(tensor<any> A, vector<int32> dims, var<bool> keepdims)->(tensor<any> B) |
-

doc/excuter/op-mem-ompsimd/list.md

@@ -13,10 +13,11 @@
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| copytensor |  none  | copytensor(tensor<any> src, tensor<any> dst)->() | T2.data = T1.data | copytensor(tensor<any> src, tensor<any> dst)->() |
+| renametensor |  none  | renametensor(tensor<any> t, var<string> new_name)->() | rename T1 to T2 | renametensor(tensor<any> t, var<string> new_name)->() |
 | newtensor |  none  | newtensor(vector<int32> shape)->(tensor<any> tensor1) | T1 =Tensor(shape=[...]) | newtensor(vector<int32> shape)->(tensor<any> tensor1) |
 | newtensor |  none  | newtensor(var<string> shape)->(tensor<any> t) | T1 =Tensor(shape=[...]) | newtensor(var<string> shape)->(tensor<any> t) |
 | deltensor |  none  | deltensor(tensor<any> t)->() | del T1 | deltensor(tensor<any> t)->() |
+| copytensor |  none  | copytensor(tensor<any> src, tensor<any> dst)->() | T2.data = T1.data | copytensor(tensor<any> src, tensor<any> dst)->() |
 
 ### io
 
@@ -25,15 +26,15 @@
 | save |  none  | save(tensor<any> t, var<string> path)->() | save(T1,path) | save(tensor<any> t, var<string> path)->() |
 | print | miaobyte | print(tensor<any> t)->() | print(T1) | print(tensor<any> t)->() |
 | print | miaobyte | print(tensor<any> t, var<string> format)->() | print(T1) | print(tensor<any> t, var<string> format)->() |
-| load |  none  | load(var<string> path)->() | load(path) | load(var<string> path)->() |
+| load |  none  | load(var<string> path)->() | mem.load(path) | load(var<string> path)->() |
 
 ### init
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| arange | miaobyte | arange(tensor<any> t, var<any> start, var<any> step)->() | arange(T1,start,step) | arange(tensor<any> t, var<any> start, var<any> step)->() |
 | normal | miaobyte | normal(tensor<any> t, var<any> mean, var<any> std, var<int32> seed)->() | normal(T1,mean,stddev,seed) | normal(tensor<any> t, var<any> mean, var<any> std, var<int32> seed)->() |
 | uniform | miaobyte | uniform(tensor<any> t, var<any> low, var<any> high, var<int32> seed)->() | uniform(T1,low,high,seed) | uniform(tensor<any> t, var<any> low, var<any> high, var<int32> seed)->() |
+| arange | miaobyte | arange(tensor<any> t, var<any> start, var<any> step)->() | arange(T1,start,step) | arange(tensor<any> t, var<any> start, var<any> step)->() |
 | constant | miaobyte | constant(tensor<any> t, var<any> value)->() | constant(T1,value) | constant(tensor<any> t, var<any> value)->() |
 
 ### elementwise
@@ -46,17 +47,17 @@
 | min | miaobyte | min(tensor<any> A, tensor<any> B)->(tensor<any> C) | T3=min(T1,T2) | min(tensor<any> A, tensor<any> B)->(tensor<any> C) |
 | maxscalar | miaobyte | maxscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=max(T1,scalar) | maxscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | divscalar | miaobyte | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=T1/scalar | divscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
-| add | cblas | add(tensor<float64|float32> a, tensor<float64|float32> b)->(tensor<float64|float32> c) | T3=T1+T2 | add(tensor<float64|float32> a, tensor<float64|float32> b)->(tensor<float64|float32> c) |
-| add | miaobyte | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
+| log | miaobyte | log(tensor<any> A)->(tensor<any> C) | T3=log(T1) | log(tensor<any> A)->(tensor<any> C) |
+| addscalar | miaobyte | addscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) | T3=T1+scalar | addscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) |
 | greater | miaobyte | greater(tensor<any> A, tensor<any> B)->(tensor<bool> mask) | mask=greater(T1,T2) | greater(tensor<any> A, tensor<any> B)->(tensor<bool> mask) |
 | lessscalar | miaobyte | lessscalar(tensor<any> A, var<any> scalar)->(tensor<bool> mask) | mask=less(T1,scalar) | lessscalar(tensor<any> A, var<any> scalar)->(tensor<bool> mask) |
 | less | miaobyte | less(tensor<any> A, tensor<any> B)->(tensor<bool> mask) | mask=less(T1,T2) | less(tensor<any> A, tensor<any> B)->(tensor<bool> mask) |
 | powscalar | miaobyte | powscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=T1^scalar | powscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | minscalar | miaobyte | minscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) | T3=min(T1,scalar) | minscalar(tensor<any> A, var<any> scalar)->(tensor<any> C) |
 | rdivscalar | miaobyte | rdivscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) | T3=scalar/T1 | rdivscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) |
 | rpowscalar | miaobyte | rpowscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) | T3=scalar^T1 | rpowscalar(var<any> scalar, tensor<any> A)->(tensor<any> C) |
-| log | miaobyte | log(tensor<any> A)->(tensor<any> C) | T3=log(T1) | log(tensor<any> A)->(tensor<any> C) |
-| addscalar | miaobyte | addscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) | T3=T1+scalar | addscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) |
+| add | cblas | add(tensor<float64|float32> a, tensor<float64|float32> b)->(tensor<float64|float32> c) | T3=T1+T2 | add(tensor<float64|float32> a, tensor<float64|float32> b)->(tensor<float64|float32> c) |
+| add | miaobyte | add(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1+T2 | add(tensor<any> a, tensor<any> b)->(tensor<any> c) |
 | sub | miaobyte | sub(tensor<any> a, tensor<any> b)->(tensor<any> c) | T3=T1-T2 | sub(tensor<any> a, tensor<any> b)->(tensor<any> c) |
 | sqrt | miaobyte | sqrt(tensor<any> A)->(tensor<any> C) | T3=sqrt(T1) | sqrt(tensor<any> A)->(tensor<any> C) |
 | subscalar | miaobyte | subscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) | T3=T1-scalar | subscalar(tensor<any> a, var<any> scalar)->(tensor<any> c) |
@@ -76,22 +77,22 @@
 | matmul | cblas | matmul(tensor<float64|float32> A, tensor<float64|float32> B)->(tensor<float64|float32> C) | T3=T1 @ T2 | matmul(tensor<float64|float32> A, tensor<float64|float32> B)->(tensor<float64|float32> C) |
 | matmul | miaobyte | matmul(tensor<any> A, tensor<any> B)->(tensor<any> C) | T3=T1 @ T2 | matmul(tensor<any> A, tensor<any> B)->(tensor<any> C) |
 
+### reduce
+
+| Operation | Author | Func Def | Math Formula | IR Instruction |
+|-----------|--------|------------|--------------|----------------|
+| prod | miaobyte | prod(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = prod(A, axis=[1 2], keepdims=false) | prod(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
+| reducemax | miaobyte | reducemax(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = reducemax(A, axis=[1 2], keepdims=false) | reducemax(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
+| sum | miaobyte | sum(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = sum(A, axis=[1 2], keepdims=false) | sum(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
+| reducemin | miaobyte | reducemin(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = reducemin(A, axis=[1 2], keepdims=false) | reducemin(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
+
 ### changeshape
 
 | Operation | Author | Func Def | Math Formula | IR Instruction |
 |-----------|--------|------------|--------------|----------------|
-| gather | miaobyte | gather(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) | T2 = T1.gather(indices=T3, axis=3) | gather(tensor<any> A, tensor<int64|int32> indices, var<int32> axis)->(tensor<any> B) |
+| indexselect | miaobyte | indexselect(tensor<any> A, tensor<int64|int32> index, var<int32> axis)->(tensor<any> B) | T2 = T1.indexselect(index=T3, axis=3) | indexselect(tensor<any> A, tensor<int64|int32> index, var<int32> axis)->(tensor<any> B) |
 | broadcastTo | miaobyte | broadcastTo(tensor<any> A, vector<int32> new_shape)->(tensor<any> B) | T2 = T1.broadcastTo(new_shape=[4,3,2]) | broadcastTo(tensor<any> A, vector<int32> new_shape)->(tensor<any> B) |
 | concat | miaobyte | concat(listtensor<any> tensors, var<int32> dim)->(tensor<any> result) | Tresult = concat([T1, T2...], axis=3) | concat(listtensor<any> tensors, var<int32> dim)->(tensor<any> result) |
 | transpose | miaobyte | transpose(tensor<any> A, vector<int32> dim_order)->(tensor<any> C) | T1.transpose(dimorder=[1,0])->T2 | transpose(tensor<any> A, vector<int32> dim_order)->(tensor<any> C) |
 | reshape | miaobyte | reshape(tensor<any> A, vector<int32> shape)->(tensor<any> B) | T1.reshape(shape)->T2 | reshape(tensor<any> A, vector<int32> shape)->(tensor<any> B) |
 
-### reduce
-
-| Operation | Author | Func Def | Math Formula | IR Instruction |
-|-----------|--------|------------|--------------|----------------|
-| reducemax | miaobyte | reducemax(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = reducemax(A, axis=[1 2], keepdims=false) | reducemax(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
-| prod | miaobyte | prod(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = prod(A, axis=[1 2], keepdims=false) | prod(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
-| sum | miaobyte | sum(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = sum(A, axis=[1 2], keepdims=false) | sum(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
-| reducemin | miaobyte | reducemin(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) | B = reducemin(A, axis=[1 2], keepdims=false) | reducemin(tensor<any> A, vector<int32> axis, var<bool> keepdims)->(tensor<any> B) |
-