From e2b19900ec37ef517d8175a7d86c1925ca9f9e91 Mon Sep 17 00:00:00 2001
From: AUTOMATIC1111 <16777216c@gmail.com>
Date: Sun, 11 Feb 2024 09:39:51 +0300
Subject: add infotext entry for emphasis; put emphasis into a separate file,
 add an option to parse but still ignore emphasis

---
 modules/sd_emphasis.py | 70 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 70 insertions(+)
 create mode 100644 modules/sd_emphasis.py

(limited to 'modules/sd_emphasis.py')

diff --git a/modules/sd_emphasis.py b/modules/sd_emphasis.py
new file mode 100644
index 00000000..654817b6
--- /dev/null
+++ b/modules/sd_emphasis.py
@@ -0,0 +1,70 @@
+from __future__ import annotations
+import torch
+
+
+class Emphasis:
+    """Emphasis class decides how to death with (emphasized:1.1) text in prompts"""
+
+    name: str = "Base"
+    description: str = ""
+
+    tokens: list[list[int]]
+    """tokens from the chunk of the prompt"""
+
+    multipliers: torch.Tensor
+    """tensor with multipliers, once for each token"""
+
+    z: torch.Tensor
+    """output of cond transformers network (CLIP)"""
+
+    def after_transformers(self):
+        """Called after cond transformers network has processed the chunk of the prompt; this function should modify self.z to apply the emphasis"""
+
+        pass
+
+
+class EmphasisNone(Emphasis):
+    name = "None"
+    description = "disable the mechanism entirely and treat (:.1.1) as literal characters"
+
+
+class EmphasisIgnore(Emphasis):
+    name = "Ignore"
+    description = "treat all empasised words as if they have no emphasis"
+
+
+class EmphasisOriginal(Emphasis):
+    name = "Original"
+    description = "the orginal emphasis implementation"
+
+    def after_transformers(self):
+        original_mean = self.z.mean()
+        self.z = self.z * self.multipliers.reshape(self.multipliers.shape + (1,)).expand(self.z.shape)
+
+        # restoring original mean is likely not correct, but it seems to work well to prevent artifacts that happen otherwise
+        new_mean = self.z.mean()
+        self.z = self.z * (original_mean / new_mean)
+
+
+class EmphasisOriginalNoNorm(EmphasisOriginal):
+    name = "No norm"
+    description = "same as orginal, but without normalization (seems to work better for SDXL)"
+
+    def after_transformers(self):
+        self.z = self.z * self.multipliers.reshape(self.multipliers.shape + (1,)).expand(self.z.shape)
+
+
+def get_current_option(emphasis_option_name):
+    return next(iter([x for x in options if x.name == emphasis_option_name]), EmphasisOriginal)
+
+
+def get_options_descriptions():
+    return ", ".join(f"{x.name}: {x.description}" for x in options)
+
+
+options = [
+    EmphasisNone,
+    EmphasisIgnore,
+    EmphasisOriginal,
+    EmphasisOriginalNoNorm,
+]
-- 
cgit v1.2.3