Adds basic layers (#80)

* adds layers

* updates

* super init

* small modification to MLP

quickvision/layers/__init__.py

@@ -0,0 +1,3 @@
+from quickvision.layers import functional
+from quickvision.layers.act_mish import Mish
+from quickvision.layers.block_mlp import MLP

quickvision/layers/act_mish.py

@@ -0,0 +1,28 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from quickvision.layers.functional import mish
+
+__all__ = ["Mish"]
+
+
+class Mish(nn.Module):
+    """
+    Applies the mish function element-wise:
+    mish(x) = x * tanh(softplus(x)) = x * tanh(ln(1 + exp(x)))
+    Shape:
+        - Input: (N, *) where * means, any number of additional
+          dimensions
+        - Output: (N, *), same shape as the input
+    Examples:
+        >>> m = Mish()
+        >>> input = torch.randn(2)
+        >>> output = m(input)
+    """
+
+    def __init__(self, inplace: bool = False):
+        super().__init__()
+        self.inplace = inplace
+
+    def forward(self, x):
+        return mish(x, inplace=self.inplace)

quickvision/layers/block_mlp.py

@@ -0,0 +1,30 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+__all__ = ["MLP"]
+
+
+class MLP(nn.Module):
+    """
+    A very simple Multi Layered Perceptron classifier.
+    Args:
+        in_features (int): Input features of Network
+        hidden_features (int): Intermediate Features of network
+        out_features (int): Output layers of network.
+    """
+
+    def __init__(self, in_features: int, hidden_features: int, proj_features: int):
+        super().__init__()
+        self.in_features = in_features
+        self.proj_features = proj_features
+        self.hidden_features = hidden_features
+        self.l1 = nn.Linear(self.in_features, self.hidden_features)
+        self.l2 = nn.Linear(self.hidden_features, self.proj_features)
+
+    def forward(self, x):
+        x = self.l1(x)
+        x = F.relu(x)
+        x = self.l2(x)
+        return (x)

quickvision/layers/functional/__init__.py

@@ -0,0 +1 @@
+from quickvision.layers.functional.act_mish import mish

quickvision/layers/functional/act_mish.py

@@ -0,0 +1,20 @@
+# Taken from Mish author himself https://github.com/digantamisra98/Mish/
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+__all__ = ["mish"]
+
+
+@torch.jit.script
+def mish(x, inplace: bool = False):
+    """
+    Applies the mish function element-wise:
+    Mish: A Self Regularized Non-Monotonic Neural Activation Function - https://arxiv.org/abs/1908.08681
+    mish(x) = x * tanh(softplus(x)) = x * tanh(ln(1 + exp(x)))
+    """
+    if inplace:
+        return x.mul_(F.tanh(F.softplus(x)))
+    else:
+        return x.mul(F.tanh(F.softplus(x)))