Update README.md

README.md

@@ -12,467 +12,541 @@ size_categories:
 - 1K<n<10K
 ---
 
-# =============================================================================
-# Ars AI Research Laboratory
-# =============================================================================
+###############################################################################
+# Ars AI Experimental Description Language (AAEDL)
+# Protocol : AAEP-0001
+# Version  : 1.0
 #
-# Protocol ID   : AAEP-0001
-# Project       : Ars Divina
-# Version       : 7.2.4a
-# Codename      : Ars Divina
-# Status        : Released
+# Project  : Ars Divina
+# Release  : 7.2.4a
+# Status   : Released
 #
-# Author        : NeverWinter13
-# Laboratory    : Ars AI
+# Laboratory : Ars AI
+# Author     : NeverWinter13
 #
 # Description
 # -----------------------------------------------------------------------------
-# Experimental Protocol Reconstruction
+# Experimental protocol reconstructed from surviving merge metadata after the
+# original laboratory notes were lost.
 #
-# This document reconstructs the complete experimental lineage of
-# Ars Divina from surviving merge metadata after the original working
-# notes were lost.
-#
-# This file is written as Python-inspired pseudocode and is intended for
-# documentation purposes only. It is NOT executable.
-#
-# =============================================================================
+# This document describes the chronological evolution of Ars Divina.
+###############################################################################
 
-from ars_ai.protocol import *
 
-# =============================================================================
-# Laboratory Specification
-# =============================================================================
+protocol "AAEP-0001" {
 
-laboratory = Laboratory(
+    project      "Ars Divina"
 
-    name         = "Ars AI",
+    codename     "Ars Divina"
 
-    researcher   = "NeverWinter13",
+    version      "7.2.4a"
 
-    philosophy   = "Explore. Analyze. Understand. Advance."
+    status       Released
 
-)
+    laboratory   "Ars AI"
 
+    researcher   "NeverWinter13"
 
-# =============================================================================
-# Runtime Environment
-# =============================================================================
+    philosophy   "Explore. Analyze. Understand. Advance."
 
-environment = Environment(
+}
 
-    platform      = "AUTOMATIC1111 WebUI",
 
-    architecture  = "SDXL",
 
-    extension     = "SuperMerger"
+###############################################################################
+# Runtime Environment
+###############################################################################
 
-)
+environment {
 
+    platform      "AUTOMATIC1111 WebUI"
 
-# =============================================================================
-# Repository
-# =============================================================================
+    extension     "SuperMerger"
 
-repository = Repository(
+    architecture  "SDXL"
 
-    checkpoints = [
+}
 
-        Checkpoint("Animagine XL 3.0"),
-        Checkpoint("Pony Diffusion v6 (Base)"),
-        Checkpoint("SDXL v1.0 Base"),
 
-        Checkpoint("Snow Pony"),
 
-        Checkpoint("IPS Mix"),
+###############################################################################
+# Repository
+###############################################################################
 
-        Checkpoint("Illustrious XL v0.1"),
-        Checkpoint("TitiMix Illustrious IL")
+repository {
 
-    ]
+    checkpoint "Animagine XL 3.0"
 
-)
+    checkpoint "Pony Diffusion v6 (Base)"
 
+    checkpoint "SDXL v1.0 Base"
 
-# =============================================================================
-# LoRA Library
-# =============================================================================
+    checkpoint "Snow Pony"
 
-lora_library = LoRALibrary(
+    checkpoint "IPS Mix"
 
-    modules = [
+    checkpoint "Illustrious XL v0.1"
 
-        LoRA(
+    checkpoint "TitiMix Illustrious IL"
 
-            name = "Anti Long Torso XL v0.1",
+}
 
-            purpose = "Structural Correction"
 
-        ),
 
-        LoRA(
+###############################################################################
+# Enhancement Library
+###############################################################################
 
-            name = "SDS_Contrast tool_XL v1.0",
+library LoRA {
 
-            purpose = "Contrast Enhancement"
+    module "Anti Long Torso XL v0.1" {
 
-        ),
+        purpose "Structural Correction"
 
-        LoRA(
+    }
 
-            name = "SPO-SDXL_4k-p_10ep_LoRA_webui",
+    module "SDS_Contrast tool_XL v1.0" {
 
-            purpose = "Detail Enhancement"
+        purpose "Contrast Enhancement"
 
-        )
+    }
 
-    ]
+    module "SPO-SDXL_4k-p_10ep_LoRA_webui" {
+
+        purpose "Detail Enhancement"
+
+    }
 
-)
+}
 
 
-# =============================================================================
+
+###############################################################################
 # Merge Profiles
-# =============================================================================
+###############################################################################
 
-MBW_PROFILE_01 = MBWProfile(
+profile MBW_PROFILE_01 {
 
-    model_a = [
+    model_a [
 
         0.0,0.0,0.0,0.0,0.0,
         0.0,0.1,0.3,0.4,0.4,
         0.4,0.4,0.4,0.3,0.1,
         0.0,0.0,0.0,0.0,0.0
 
-    ],
+    ]
 
-    model_b = [
+    model_b [
 
         1.0,1.0,1.0,1.0,1.0,
         1.0,0.9,0.7,0.6,0.6,
         0.6,0.6,0.6,0.7,0.9,
         1.0,1.0,1.0,1.0,1.0
 
-    ],
+    ]
+
+    elemental {
 
-    elemental = {
+        "BASE:conditioner.embedders.0" = 0.25
 
-        "BASE:conditioner.embedders.0" : 0.25,
-        "BASE:conditioner.embedders.1" : 0.35,
-        "BASE"                         : 0.00
+        "BASE:conditioner.embedders.1" = 0.35
+
+        "BASE" = 0.00
 
     }
 
-)
+}
+
 
 
-MBW_PROFILE_02 = MBWProfile(
+profile MBW_PROFILE_02 {
 
-    model_a = [
+    model_a [
 
         0.0,0.0,0.0,0.0,0.0,
         0.0,0.0,0.3,0.4,0.4,
         0.4,0.4,0.4,0.4,0.4,
         0.3,0.2,0.0,0.0,0.0
 
-    ],
+    ]
 
-    model_b = [
+    model_b [
 
         1.0,1.0,1.0,1.0,1.0,
         1.0,1.0,0.7,0.6,0.6,
         0.6,0.6,0.6,0.6,0.6,
         0.7,0.8,1.0,1.0,1.0
 
-    ],
+    ]
+
+    elemental {
+
+        "BASE:conditioner.embedders.0" = 0.20
 
-    elemental = {
+        "BASE:conditioner.embedders.1" = 0.30
 
-        "BASE:conditioner.embedders.0" : 0.20,
-        "BASE:conditioner.embedders.1" : 0.30,
-        "BASE"                         : 0.00
+        "BASE" = 0.00
 
     }
 
-)
+}
 
 
-MBW_PROFILE_03 = MBWProfile(
 
-    model_a = [
+profile MBW_PROFILE_03 {
+
+    model_a [
 
         0.0,0.0,0.0,0.0,0.0,
         0.0,0.0,0.4,0.4,0.4,
         0.3,0.4,0.4,0.4,0.4,
         0.0,0.0,0.0,0.0,0.0
 
-    ],
+    ]
 
-    model_b = [
+    model_b [
 
         1.0,1.0,1.0,1.0,1.0,
         1.0,1.0,0.6,0.6,0.6,
         0.7,0.6,0.6,0.6,0.6,
         1.0,1.0,1.0,1.0,1.0
 
-    ],
+    ]
 
-    elemental = {
+    elemental {
 
-        "BASE" : 0.00
+        "BASE" = 0.00
 
     }
 
-)
+}
 
 
-# =============================================================================
-# Experimental Protocol
-# =============================================================================
-#
-# EXP_001–005 : Foundation Construction
-# EXP_006–009 : Cross-Architecture Integration
-# EXP_010–013 : LoRA Enhancement
-# EXP_014–015 : Final Consolidation
-#
-# =============================================================================
 
+###############################################################################
+# Experimental Workflow
+###############################################################################
 
-# -----------------------------------------------------------------------------
-# Phase I — Foundation Construction
-# -----------------------------------------------------------------------------
+phase "Foundation Construction" {
 
-EXP_001 = merge(
+    experiment EXP_001 {
 
-    input_a = Checkpoint("Animagine XL 3.0") @ 0.45,
-    input_b = Checkpoint("Pony Diffusion v6 (Base)") @ 0.55,
+        merge {
 
-    operation = WeightSum.CosineB
+            model_a    "Animagine XL 3.0"
+            weight_a   0.45
 
-)
+            model_b    "Pony Diffusion v6 (Base)"
+            weight_b   0.55
 
+            operation  WeightSum.CosineB
 
-EXP_002 = merge(
+        }
 
-    input_a = EXP_001 @ 0.45,
-    input_b = Checkpoint("SDXL v1.0 Base") @ 0.55,
+    }
 
-    operation = AddDifference.TrainDifference
 
-)
+    experiment EXP_002 {
 
+        merge {
 
-EXP_003 = merge(
+            model_a    EXP_001
+            weight_a   0.45
 
-    input_a = EXP_002 @ 0.65,
-    input_b = Checkpoint("Snow Pony") @ 0.35,
+            model_b    "SDXL v1.0 Base"
+            weight_b   0.55
 
-    operation = AddDifference.TrainDifference
+            operation  AddDifference.TrainDifference
 
-)
+        }
 
+    }
 
-EXP_004 = merge(
 
-    input_a = EXP_003 @ 0.45,
-    input_b = EXP_002 @ 0.55,
+    experiment EXP_003 {
 
-    operation = AddDifference.TrainDifference
+        merge {
 
-)
+            model_a    EXP_002
+            weight_a   0.65
 
+            model_b    "Snow Pony"
+            weight_b   0.35
 
-EXP_005 = merge(
+            operation  AddDifference.TrainDifference
 
-    input_a = EXP_004 @ 0.95,
-    input_b = Checkpoint("IPS Mix") @ 0.05,
+        }
 
-    operation = WeightSum.CosineA
+    }
 
-)
 
+    experiment EXP_004 {
 
-# -----------------------------------------------------------------------------
-# Phase II — Cross-Architecture Integration
-# -----------------------------------------------------------------------------
+        merge {
 
-EXP_006 = merge(
+            model_a    EXP_003
+            weight_a   0.45
 
-    input_a = EXP_005,
+            model_b    EXP_002
+            weight_b   0.55
 
-    input_b = Checkpoint("Illustrious XL v0.1"),
+            operation  AddDifference.TrainDifference
 
-    operation = WeightSum.Normal,
+        }
 
-    profile = MBW_PROFILE_01
+    }
 
-)
 
+    experiment EXP_005 {
 
-EXP_007 = merge(
+        merge {
 
-    input_a = EXP_006,
+            model_a    EXP_004
+            weight_a   0.95
 
-    input_b = EXP_005,
+            model_b    "IPS Mix"
+            weight_b   0.05
 
-    operation = WeightSum.CosineA,
+            operation  WeightSum.CosineA
 
-    profile = MBW_PROFILE_01
+        }
 
-)
+    }
 
+}
 
-EXP_008 = merge(
 
-    input_a = EXP_007,
 
-    input_b = Checkpoint("TitiMix Illustrious IL"),
+phase "Cross-Architecture Integration" {
 
-    operation = AddDifference.TrainDifference,
+    experiment EXP_006 {
 
-    profile = MBW_PROFILE_01
+        merge {
 
-)
+            model_a   EXP_005
 
+            model_b   "Illustrious XL v0.1"
 
-EXP_009 = merge(
+            operation WeightSum.Normal
 
-    input_a = EXP_008,
+            profile   MBW_PROFILE_01
 
-    input_b = EXP_007,
+        }
 
-    operation = WeightSum.CosineA,
+    }
 
-    profile = MBW_PROFILE_02
 
-)
+    experiment EXP_007 {
 
+        merge {
 
-# -----------------------------------------------------------------------------
-# Phase III — LoRA Enhancement
-# -----------------------------------------------------------------------------
+            model_a   EXP_006
+
+            model_b   EXP_005
 
-EXP_010 = EXP_009.bake(
+            operation WeightSum.CosineA
 
-    LoRA("Anti Long Torso XL v0.1") @ 4.0
+            profile   MBW_PROFILE_01
 
-)
+        }
 
+    }
 
-EXP_011 = merge(
 
-    input_a = EXP_010,
+    experiment EXP_008 {
 
-    input_b = EXP_009,
+        merge {
 
-    operation = WeightSum.Normal,
+            model_a   EXP_007
 
-    profile = MBW_PROFILE_02
+            model_b   "TitiMix Illustrious IL"
 
-)
+            operation AddDifference.TrainDifference
 
+            profile   MBW_PROFILE_01
 
-EXP_012 = EXP_011.bake(
+        }
 
-    LoRA("SDS_Contrast tool_XL v1.0") @ 0.25
+    }
 
-)
 
+    experiment EXP_009 {
 
-EXP_013 = EXP_012.bake(
+        merge {
 
-    LoRA("SPO-SDXL_4k-p_10ep_LoRA_webui") @ 1.0
+            model_a   EXP_008
 
-)
+            model_b   EXP_007
 
+            operation WeightSum.CosineA
 
-# -----------------------------------------------------------------------------
-# Phase IV — Final Consolidation
-# -----------------------------------------------------------------------------
+            profile   MBW_PROFILE_02
 
-EXP_014 = merge(
+        }
 
-    input_a = EXP_013,
+    }
 
-    input_b = EXP_011,
+}
 
-    operation = AddDifference.TrainDifference,
 
-    profile = MBW_PROFILE_02
 
-)
+phase "LoRA Enhancement" {
 
+    experiment EXP_010 {
 
-EXP_015 = merge(
+        bake {
 
-    input_a = EXP_014,
+            source     EXP_009
 
-    input_b = EXP_013,
+            lora       "Anti Long Torso XL v0.1"
 
-    operation = WeightSum.Normal,
+            strength   4.0
 
-    profile = MBW_PROFILE_03
+        }
 
-)
+    }
 
 
-# =============================================================================
-# Published Artifact
-# =============================================================================
+    experiment EXP_011 {
 
-ars_divina = Artifact(
+        merge {
 
-    source = EXP_015,
+            model_a   EXP_010
 
-    name = "Ars Divina",
+            model_b   EXP_009
 
-    version = "7.2.4a",
+            operation WeightSum.Normal
 
-    laboratory = laboratory,
+            profile   MBW_PROFILE_02
 
-    status = Released
+        }
+
+    }
 
-)
 
+    experiment EXP_012 {
 
-# =============================================================================
-# Experimental Notes
-# =============================================================================
+        bake {
 
-Notes("""
+            source     EXP_011
 
-Ars Divina was not created through a single merge operation.
+            lora       "SDS_Contrast tool_XL v1.0"
 
-Each experiment concluded with qualitative inference using a fixed prompt
-suite before restarting the WebUI to generate X/Y/Z comparison plots
-under a static seed (1111).
+            strength   0.25
 
-Only after evaluating the resulting checkpoint was the next experiment
-performed.
+        }
 
-Consequently, this protocol documents the chronological evolution of
-Ars Divina rather than an optimized merge recipe.
+    }
 
-Every experiment generated observations.
 
-Every checkpoint became another hypothesis.
+    experiment EXP_013 {
 
-Every merge inspired another experiment.
+        bake {
 
-""")
+            source     EXP_012
 
+            lora       "SPO-SDXL_4k-p_10ep_LoRA_webui"
 
+            strength   1.0
 
-# =============================================================================
-# Discussion
-# =============================================================================
+        }
+
+    }
+
+}
+
+
+
+phase "Final Consolidation" {
+
+    experiment EXP_014 {
+
+        merge {
+
+            model_a   EXP_013
+
+            model_b   EXP_011
+
+            operation AddDifference.TrainDifference
+
+            profile   MBW_PROFILE_02
+
+        }
+
+    }
+
+
+    experiment EXP_015 {
+
+        merge {
+
+            model_a   EXP_014
+
+            model_b   EXP_013
+
+            operation WeightSum.Normal
+
+            profile   MBW_PROFILE_03
+
+        }
+
+    }
+
+}
+
+
+
+###############################################################################
+# Published Artifact
+###############################################################################
+
+artifact "Ars Divina" {
+
+    source       EXP_015
+
+    version      "7.2.4a"
+
+    laboratory   "Ars AI"
+
+    state        Released
+
+}
+
+
+
+###############################################################################
+# Evaluation Protocol
+###############################################################################
+
+evaluation {
+
+    after_each_experiment {
+
+        inference
+
+        restart_webui
+
+        xyz_plot {
+
+            seed 1111
+
+        }
+
+        qualitative_review
+
+        document_results
+
+    }
 
-Discussion(
+}
 
-    questions   = True,
 
-    feedback    = True,
 
-    suggestions = True
+###############################################################################
+# End of Protocol
+###############################################################################
 
-)
+###############################################################################
+# FAQs
+###############################################################################
+?