ncnn/docs/developer-guide/operators.md

* [absval](#absval)
* [argmax](#argmax)
* [batchnorm](#batchnorm)
* [bias](#bias)
* [binaryop](#binaryop)
* [bnll](#bnll)
* [cast](#cast)
* [clip](#clip)
* [concat](#concat)
* [convolution](#convolution)
* [convolutiondepthwise](#convolutiondepthwise)
* [crop](#crop)
* [deconvolution](#deconvolution)
* [deconvolutiondepthwise](#deconvolutiondepthwise)
* [dequantize](#dequantize)
* [dropout](#dropout)
* [eltwise](#eltwise)
* [elu](#elu)
* [exp](#exp)
* [flatten](#flatten)
* [gelu](#gelu)
* [gemm](#gemm)
* [groupnorm](#groupnorm)
* [gru](#gru)
* [hardsigmoid](#hardsigmoid)
* [hardswish](#hardswish)
* [innerproduct](#innerproduct)
* [input](#input)
* [instancenorm](#instancenorm)
* [interp](#interp)
* [layernorm](#layernorm)
* [log](#log)
* [lrn](#lrn)
* [lstm](#lstm)
* [memorydata](#memorydata)
* [mish](#mish)
* [multiheadattention](#multiheadattention)
* [pooling](#pooling)
* [selu](#selu)
* [sigmoid](#sigmoid)
* [slice](#slice)
* [softmax](#softmax)
* [softplus](#softplus)
* [split](#split)
* [swish](#swish)
* [tanh](#tanh)
* [threshold](#threshold)
* [unaryop](#unaryop)

# absval
```
y = abs(x)
```

* one_blob_only
* support_inplace

# argmax
```
y = argmax(x, out_max_val, topk)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | out_max_val   | int   | 0         |                   |
| 1         | topk          | int   | 1         |                   |

# batchnorm
```
y = (x - mean) / sqrt(var + eps) * slope + bias
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | channels      | int   | 0         |                   |
| 1         | eps           | float | 0.f       |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| slope_data    | float | [channels]            |
| mean_data     | float | [channels]            |
| var_data      | float | [channels]            |
| bias_data     | float | [channels]            |

# bias
```
y = x + bias
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | bias_data_size| int   | 0         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| bias_data     | float | [channels]            |

# binaryop
 This operation is used for binary computation, and the calculation rule depends on the [broadcasting rule](https://github.com/Tencent/ncnn/wiki/binaryop-broadcasting).
```
C = binaryop(A, B)
```
if with_scalar = 1:
- one_blob_only
- support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | op_type       | int   | 0         | Operation type as follows |
| 1         | with_scalar   | int   | 0         | with_scalar=0 B is a matrix, with_scalar=1 B is a scalar |
| 2         | b             | float | 0.f       | When B is a scalar, B = b |

Operation type:
- 0 = ADD
- 1 = SUB
- 2 = MUL
- 3 = DIV
- 4 = MAX
- 5 = MIN
- 6 = POW
- 7 = RSUB
- 8 = RDIV

# bnll
```
y = log(1 + e^(-x)) , x > 0
y = log(1 + e^x),     x < 0
```

* one_blob_only
* support_inplace

# cast
```
y = cast(x)
```

* one_blob_only
* support_packing

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | type_from     | int   | 0         |                   |
| 1         | type_to       | int   | 0         |                   |

Element type:

- 0 = auto
- 1 = float32
- 2 = float16
- 3 = int8
- 4 = bfloat16

# clip
```
y = clamp(x, min, max)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | min           | float | -FLT_MAX  |                   |
| 1         | max           | float | FLT_MAX   |                   |

# concat
```
y = concat(x0, x1, x2, ...) by axis
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | axis          | int   | 0         |                   |

# convolution
```
x2 = pad(x, pads, pad_value)
x3 = conv(x2, weight, kernel, stride, dilation) + bias
y = activation(x3, act_type, act_params)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         |                   |
| 1         | kernel_w      | int   | 0         |                   |
| 2         | dilation_w    | int   | 1         |                   |
| 3         | stride_w      | int   | 1         |                   |
| 4         | pad_left      | int   | 0         |                   |
| 5         | bias_term     | int   | 0         |                   |
| 6         | weight_data_size| int | 0         |                   |
| 8         | int8_scale_term| int  | 0         |                   |
| 9         | activation_type| int  | 0         |                   |
| 10        | activation_params| array | [ ]    |                   |
| 11        | kernel_h      | int   | kernel_w  |                   |
| 12        | dilation_h    | int   | dilation_w |                  |
| 13        | stride_h      | int   | stride_w  |                   |
| 15        | pad_right     | int   | pad_left  |                   |
| 14        | pad_top       | int   | pad_left  |                   |
| 16        | pad_bottom    | int   | pad_top   |                   |
| 18        | pad_value     | float | 0.f       |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_data   | float/fp16/int8 | [kernel_w, kernel_h, num_input, num_output] |
| bias_data     | float | [num_output]          |
| weight_data_int8_scales| float | [num_output] |
| bottom_blob_int8_scales| float | [1]          |
| top_blob_int8_scales| float | [1]             |

# convolutiondepthwise
```
x2 = pad(x, pads, pad_value)
x3 = conv(x2, weight, kernel, stride, dilation, group) + bias
y = activation(x3, act_type, act_params)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         |                   |
| 1         | kernel_w      | int   | 0         |                   |
| 2         | dilation_w    | int   | 1         |                   |
| 3         | stride_w      | int   | 1         |                   |
| 4         | pad_left      | int   | 0         |                   |
| 5         | bias_term     | int   | 0         |                   |
| 6         | weight_data_size| int | 0         |                   |
| 7         | group         | int   | 1         |                   |
| 8         | int8_scale_term| int  | 0         |                   |
| 9         | activation_type| int  | 0         |                   |
| 10        | activation_params| array | [ ]    |                   |
| 11        | kernel_h      | int   | kernel_w  |                   |
| 12        | dilation_h    | int   | dilation_w |                  |
| 13        | stride_h      | int   | stride_w  |                   |
| 15        | pad_right     | int   | pad_left  |                   |
| 14        | pad_top       | int   | pad_left  |                   |
| 16        | pad_bottom    | int   | pad_top   |                   |
| 18        | pad_value     | float | 0.f       |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_data   | float/fp16/int8 | [kernel_w, kernel_h, num_input / group, num_output / group, group] |
| bias_data     | float | [num_output]          |
| weight_data_int8_scales| float | [group]      |
| bottom_blob_int8_scales| float | [1]          |
| top_blob_int8_scales| float | [1]             |

# crop
```
y = crop(x)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | woffset       | int   | 0         |                   |
| 1         | hoffset       | int   | 0         |                   |
| 2         | coffset       | int   | 1         |                   |
| 3         | outw          | int   | 1         |                   |
| 4         | outh          | int   | 0         |                   |
| 5         | outc          | int   | 0         |                   |
| 6         | woffset2      | int   | 0         |                   |
| 7         | hoffset2      | int   | 1         |                   |
| 8         | coffset2      | int   | 0         |                   |
| 9         | starts        | array | [ ]       |                   |
| 10        | ends          | array | [ ]       |                   |
| 11        | axes          | array | [ ]       |                   |

# deconvolution
```
x2 = deconv(x, weight, kernel, stride, dilation) + bias
x3 = depad(x2, pads, pad_value)
y = activation(x3, act_type, act_params)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         |                   |
| 1         | kernel_w      | int   | 0         |                   |
| 2         | dilation_w    | int   | 1         |                   |
| 3         | stride_w      | int   | 1         |                   |
| 4         | pad_left      | int   | 0         |                   |
| 5         | bias_term     | int   | 0         |                   |
| 6         | weight_data_size| int | 0         |                   |
| 8         | int8_scale_term| int  | 0         |                   |
| 9         | activation_type| int  | 0         |                   |
| 10        | activation_params| array | [ ]    |                   |
| 11        | kernel_h      | int   | kernel_w  |                   |
| 12        | dilation_h    | int   | dilation_w |                  |
| 13        | stride_h      | int   | stride_w  |                   |
| 15        | pad_right     | int   | pad_left  |                   |
| 14        | pad_top       | int   | pad_left  |                   |
| 16        | pad_bottom    | int   | pad_top   |                   |
| 18        | pad_value     | float | 0.f       |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_data   | float/fp16/int8 | [kernel_w, kernel_h, num_input, num_output] |
| bias_data     | float | [num_output]          |

# deconvolutiondepthwise
```
x2 = deconv(x, weight, kernel, stride, dilation, group) + bias
x3 = depad(x2, pads, pad_value)
y = activation(x3, act_type, act_params)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         |                   |
| 1         | kernel_w      | int   | 0         |                   |
| 2         | dilation_w    | int   | 1         |                   |
| 3         | stride_w      | int   | 1         |                   |
| 4         | pad_left      | int   | 0         |                   |
| 5         | bias_term     | int   | 0         |                   |
| 6         | weight_data_size| int | 0         |                   |
| 7         | group         | int   | 1         |                   |
| 8         | int8_scale_term| int  | 0         |                   |
| 9         | activation_type| int  | 0         |                   |
| 10        | activation_params| array | [ ]    |                   |
| 11        | kernel_h      | int   | kernel_w  |                   |
| 12        | dilation_h    | int   | dilation_w |                  |
| 13        | stride_h      | int   | stride_w  |                   |
| 15        | pad_right     | int   | pad_left  |                   |
| 14        | pad_top       | int   | pad_left  |                   |
| 16        | pad_bottom    | int   | pad_top   |                   |
| 18        | pad_value     | float | 0.f       |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_data   | float/fp16/int8 | [kernel_w, kernel_h, num_input / group, num_output / group, group] |
| bias_data     | float | [num_output]          |

# dequantize
```
y = x * scale + bias
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | scale         | float | 1.f       |                   |
| 1         | bias_term     | int   | 0         |                   |
| 2         | bias_data_size| int   | 0         |                   |

# dropout
```
y = x * scale
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | scale         | float | 1.f       |                   |

# eltwise

```
y = elementwise_op(x0, x1, ...)
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | op_type       | int   | 0         |                   |
| 1         | coeffs        | array | [ ]       |                   |

Operation type:
- 0 = PROD
- 1 = SUM
- 2 = MAX

# elu
```
if x < 0    y = (exp(x) - 1) * alpha
else        y = x
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | alpha         | float | 0.1f      |                   |

# exp
```
if base == -1   y = exp(shift + x * scale)
else            y = pow(base, (shift + x * scale))
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | base          | float | -1.f      |                   |
| 1         | scale         | float | 1.f       |                   |
| 2         | shift         | float | 0.f       |                   |

# flatten
Reshape blob to 1 dimension

* one_blob_only

# gelu
```
if fast_gelu == 1   y = 0.5 * x * (1 + tanh(0.79788452 * (x + 0.044715 * x * x * x)));
else                y = 0.5 * x * erfc(-0.70710678 * x)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | fast_gelu     | int   | 0         | use approximation |

# gemm
```
a = transA ? transpose(x0) : x0
b = transb ? transpose(x1) : x1
c = x2
y = gemm(a, b) * alpha + c * beta
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | alpha         | float | 1.f       |                   |
| 1         | beta          | float | 1.f       |                   |
| 2         | transA        | int   | 0         |                   |
| 3         | transb        | int   | 0         |                   |

# groupnorm
```
split x along channel axis into group x0, x1 ...
l2 normalize for each group x0, x1 ...
y = x * gamma + beta
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | group         | int   | 1         |                   |
| 1         | channels      | int   | 0         |                   |
| 2         | eps           | float | 0.001f    | x = x / sqrt(var + eps) |
| 3         | affine        | int   | 1         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| gamma_data    | float | [channels]            |
| beta_data     | float | [channels]            |

# gru
Apply a single-layer GRU to a feature sequence of `T` timesteps. The input blob shape is `[w=input_size, h=T]` and the output blob shape is `[w=num_output, h=T]`.

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         | hidden size of output |
| 1         | weight_data_size| int | 0         | total size of weight matrix |
| 2         | direction     | int   | 0         | 0=forward, 1=reverse, 2=bidirectional |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_xc_data| float | [input_size, num_output * 3, num_directions] |
| bias_c_data   | float | [num_output, 4, num_directions] |
| weight_hc_data| float | [num_output, num_output * 3, num_directions] |

# hardsigmoid
```
y = clamp(x * alpha + beta, 0, 1)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | alpha         | float | 0.2f      |                   |
| 1         | beta          | float | 0.5f      |                   |

# hardswish
```
y = x * clamp(x * alpha + beta, 0, 1)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | alpha         | float | 0.2f      |                   |
| 1         | beta          | float | 0.5f      |                   |

# innerproduct
```
x2 = innerproduct(x, weight) + bias
y = activation(x2, act_type, act_params)
```

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         |                   |
| 1         | bias_term     | int   | 0         |                   |
| 2         | weight_data_size| int | 0         |                   |
| 8         | int8_scale_term| int  | 0         |                   |
| 9         | activation_type| int  | 0         |                   |
| 10        | activation_params| array | [ ]    |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_data   | float/fp16/int8 | [num_input, num_output] |
| bias_data     | float | [num_output]          |
| weight_data_int8_scales| float | [num_output] |
| bottom_blob_int8_scales| float | [1]          |

# input
```
y = input
```

* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | w             | int   | 0         |                   |
| 1         | h             | int   | 0         |                   |
| 2         | c             | int   | 0         |                   |

# instancenorm
```
split x along channel axis into instance x0, x1 ...
l2 normalize for each channel instance x0, x1 ...
y = x * gamma + beta
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | channels      | int   | 0         |                   |
| 1         | eps           | float | 0.001f    | x = x / sqrt(var + eps) |
| 2         | affine        | int   | 1         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| gamma_data    | float | [channels]            |
| beta_data     | float | [channels]            |

# interp
```
if dynamic_target_size == 0     y = resize(x) by fixed size or scale
else                            y = resize(x0, size(x1))
```

* one_blob_only if dynamic_target_size == 0

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | resize_type   | int   | 0         |                   |
| 1         | height_scale  | float | 1.f       |                   |
| 2         | width_scale   | float | 1.f       |                   |
| 3         | output_height | int   | 0         |                   |
| 4         | output_width  | int   | 0         |                   |
| 5         | dynamic_target_size| int | 0      |                   |
| 6         | align_corner  | int   | 0         |                   |

Resize type:
- 1 = Nearest
- 2 = Bilinear
- 3 = Bicubic

# layernorm
```
split x along outmost axis into part x0, x1 ...
l2 normalize for each part x0, x1 ...
y = x * gamma + beta by elementwise
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | affine_size   | int   | 0         |                   |
| 1         | eps           | float | 0.001f    | x = x / sqrt(var + eps) |
| 2         | affine        | int   | 1         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| gamma_data    | float | [affine_size]         |
| beta_data     | float | [affine_size]         |

# log
```
if base == -1   y = log(shift + x * scale)
else            y = log(shift + x * scale) / log(base)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | base          | float | -1.f      |                   |
| 1         | scale         | float | 1.f       |                   |
| 2         | shift         | float | 0.f       |                   |

# lrn
```
if region_type == ACROSS_CHANNELS   square_sum = sum of channel window of local_size
if region_type == WITHIN_CHANNEL    square_sum = sum of spatial window of local_size
y = x * pow(bias + alpha * square_sum / (local_size * local_size), -beta)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | region_type   | int   | 0         |                   |
| 1         | local_size    | int   | 5         |                   |
| 2         | alpha         | float | 1.f       |                   |
| 3         | beta          | float | 0.75f     |                   |
| 4         | bias          | float | 1.f       |                   |

Region type:
- 0 = ACROSS_CHANNELS
- 1 = WITHIN_CHANNEL

# lstm
Apply a single-layer LSTM to a feature sequence of `T` timesteps. The input blob shape is `[w=input_size, h=T]` and the output blob shape is `[w=num_output, h=T]`.

* one_blob_only

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | num_output    | int   | 0         | hidden size of output |
| 1         | weight_data_size| int | 0         | total size of IFOG weight matrix |
| 2         | direction     | int   | 0         | 0=forward, 1=reverse, 2=bidirectional |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| weight_xc_data| float | [input_size, num_output * 4, num_directions] |
| bias_c_data   | float | [num_output, 4, num_directions] |
| weight_hc_data| float | [num_output, num_output * 4, num_directions] |

# memorydata
```
y = data
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | w             | int   | 0         |                   |
| 1         | h             | int   | 0         |                   |
| 2         | c             | int   | 0         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| data          | float | [w, h, c]             |

# mish
```
y = x * tanh(log(exp(x) + 1))
```

* one_blob_only
* support_inplace

# multiheadattention
```
split q k v into num_head part q0, k0, v0, q1, k1, v1 ...
for each num_head part
    xq = affine(q) / (embed_dim / num_head)
    xk = affine(k)
    xv = affine(v)
    xqk = xq * xk
    softmax_inplace(xqk)
    xqkv = xqk * xv
    merge xqkv to out
y = affine(out)
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | embed_dim     | int   | 0         |                   |
| 1         | num_head      | int   | 1         |                   |
| 2         | weight_data_size| int | 0         |                   |

| weight        | type  | shape                 |
| ------------- | ----- | --------------------- |
| q_weight_data | float/fp16/int8 | [weight_data_size] |
| q_bias_data   | float | [embed_dim]           |
| k_weight_data | float/fp16/int8 | [weight_data_size] |
| k_bias_data   | float | [embed_dim]           |
| v_weight_data | float/fp16/int8 | [weight_data_size] |
| v_bias_data   | float | [embed_dim]           |
| out_weight_data| float/fp16/int8 | [weight_data_size] |
| out_bias_data | float | [embed_dim]           |

# pooling

```
x2 = pad(x, pads)
x3 = pooling(x2, kernel, stride)
```

| param id  | name           | type | default   | description       |
| --------- | -------------- | ---- | --------- | ----------------- |
| 0         | pooling_type   | int  | 0         |                   |
| 1         | kernel_w       | int  | 0         |                   |
| 2         | stride_w       | int  | 1         |                   |
| 3         | pad_left       | int  | 0         |                   |
| 4         | global_pooling | int  | 0         |                   |
| 5         | pad_mode       | int  | 0         |                   |
| 11        | kernel_h       | int  | kernel_w  |                   |
| 12        | stride_h       | int  | stride_w  |                   |
| 13        | pad_top        | int  | pad_left  |                   |
| 14        | pad_right      | int  | pad_left  |                   |
| 15        | pad_bottom     | int  | pad_top   |                   |

Pooling type:
- 0 = MAX
- 1 = AVG

Pad mode:
- 0 = full padding
- 1 = valid padding
- 2 = tensorflow padding=SAME or onnx padding=SAME_UPPER
- 3 = onnx padding=SAME_LOWER

# selu
```
if x < 0    y = (exp(x) - 1.f) * alpha * lambda
else        y = x * lambda
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | alpha         | float | 1.67326324f|                  |
| 1         | lambda        | float | 1.050700987f|                 |

# sigmoid
```
y = 1 / (1 + exp(-x))
```

* one_blob_only
* support_inplace

# slice
```
split x along axis into slices, each part slice size is based on slices array
```

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | slices        | array | [ ]       |                   |
| 1         | axis          | int   | 0         |                   |

# softmax
```
softmax(x, axis)
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | axis          | int   | 0         |                   |
| 1         | fixbug0       | int   | 0         | hack for bug fix, should be 1 |

# softplus
```
y = log(exp(x) + 1)
```

* one_blob_only
* support_inplace

# split
```
y0, y1 ... = x
```

# swish
```
y = x / (1 + exp(-x))
```

* one_blob_only
* support_inplace

# tanh
```
y = tanh(x)
```

* one_blob_only
* support_inplace

# threshold
```
if x > threshold    y = 1
else                y = 0
```

* one_blob_only
* support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | threshold     | float | 0.f       |                   |

# unaryop
```
y = unaryop(x)
```

- one_blob_only
- support_inplace

| param id  | name          | type  | default   | description       |
| --------- | ------------- | ----- | --------- | ----------------- |
| 0         | op_type       | int   | 0         | Operation type as follows |

Operation type:
- 0 = ABS
- 1 = NEG
- 2 = FLOOR
- 3 = CEIL
- 4 = SQUARE
- 5 = SQRT
- 6 = RSQ
- 7 = EXP
- 8 = LOG
- 9 = SIN
- 10 = COS
- 11 = TAN
- 12 = ASIN
- 13 = ACOS
- 14 = ATAN
- 15 = RECIPROCAL
- 16 = TANH