From 2bcb14def38fda5f91f563cfb598e43aaa666135 Mon Sep 17 00:00:00 2001
From: David Rotermund <54365609+davrot@users.noreply.github.com>
Date: Sun, 23 Jul 2023 01:54:23 +0200
Subject: [PATCH] Add files via upload

---
 functions/SoftmaxPower.py | 29 ++++++++++++++++++++++++++---
 1 file changed, 26 insertions(+), 3 deletions(-)

diff --git a/functions/SoftmaxPower.py b/functions/SoftmaxPower.py
index 22b97cf..8a39ddc 100644
--- a/functions/SoftmaxPower.py
+++ b/functions/SoftmaxPower.py
@@ -5,14 +5,20 @@ class SoftmaxPower(torch.nn.Module):
     dim: int | None
     power: float
     mean_mode: bool
+    no_input_mode: bool
 
     def __init__(
-        self, power: float = 2.0, dim: int | None = None, mean_mode: bool = False
+        self,
+        power: float = 2.0,
+        dim: int | None = None,
+        mean_mode: bool = False,
+        no_input_mode: bool = False,
     ) -> None:
         super().__init__()
         self.dim = dim
         self.power = power
         self.mean_mode = mean_mode
+        self.no_input_mode = no_input_mode
 
     def __setstate__(self, state):
         super().__setstate__(state)
@@ -22,6 +28,8 @@ class SoftmaxPower(torch.nn.Module):
             self.power = 2.0
         if not hasattr(self, "mean_mode"):
             self.mean_mode = False
+        if not hasattr(self, "no_input_mode"):
+            self.no_input_mode = False
 
     def forward(self, input: torch.Tensor) -> torch.Tensor:
         output: torch.Tensor = torch.abs(input).pow(self.power)
@@ -30,10 +38,25 @@ class SoftmaxPower(torch.nn.Module):
         else:
             output = output / output.sum(dim=self.dim, keepdim=True)
 
-        if self.mean_mode:
+        if self.no_input_mode:
+            return output
+        elif self.mean_mode:
             return torch.abs(input).mean(dim=1, keepdim=True) * output
         else:
             return input * output
 
     def extra_repr(self) -> str:
-        return f"dim={self.dim} ; power={self.power}"
+        if self.power != 0.0:
+            return (
+                f"dim={self.dim}; "
+                f"power={self.power}; "
+                f"mean_mode={self.mean_mode}; "
+                f"no_input_mode={self.no_input_mode}"
+            )
+        else:
+            return (
+                f"dim={self.dim}; "
+                "exp-mode; "
+                f"mean_mode={self.mean_mode}; "
+                f"no_input_mode={self.no_input_mode}"
+            )