PyGAAMAS: evaluate Qwen3 in the investment game

README.md

@@ -52,7 +52,7 @@ Figure below highlight significant differences in decision-making
 consistency among the evaluated models. <tt>GPT-4.5</tt>, <tt>LLama3.3:latest</tt> 
 and <tt>DeepSeek-R1:7b</tt> stand out with a
 perfect CCEI score of 1.0, indicating flawless rationality in decision-making.
-<tt>Mistral-Small</tt> and <tt>Mixtral:8x7b</tt> demonstrate the next highest level of rationality. 
+<tt>Qwen3</tt>, <tt>Mistral-Small</tt> and <tt>Mixtral:8x7b</tt> demonstrate the next highest level of rationality. 
 <tt>Llama3</tt> performs moderately well, with CCEI values ranging between 0.2 and 0.74. 
 <tt>DeepSeek-R1</tt> exhibits
 inconsistent behavior, with CCEI scores varying widely between 0.15 and 0.83.

data/investment/investment.csv

@@ -234,4 +234,34 @@ iteration,model,temperature,ccei
 28,gpt-4.5-preview-2025-02-27,0.0,1.0
 29,gpt-4.5-preview-2025-02-27,0.0,1.0
 30,gpt-4.5-preview-2025-02-27,0.0,1.0
+1,qwen3,0.0,1.0
+2,qwen3,0.0,1.0
+3,qwen3,0.0,1.0
+4,qwen3,0.0,1.0
+5,qwen3,0.0,1.0
+6,qwen3,0.0,0.85
+7,qwen3,0.0,1.0
+8,qwen3,0.0,1.0
+9,qwen3,0.0,1.0
+10,qwen3,0.0,1.0
+11,qwen3,0.0,1.0
+12,qwen3,0.0,1.0
+13,qwen3,0.0,1.0
+14,qwen3,0.0,1.0
+15,qwen3,0.0,1.0
+16,qwen3,0.0,1.0
+17,qwen3,0.0,1.0
+18,qwen3,0.0,1.0
+19,qwen3,0.0,1.0
+20,qwen3,0.0,1.0
+21,qwen3,0.0,0.85
+22,qwen3,0.0,1.0
+23,qwen3,0.0,1.0
+24,qwen3,0.0,1.0
+25,qwen3,0.0,1.0
+26,qwen3,0.0,1.0
+27,qwen3,0.0,1.0
+28,qwen3,0.0,0.85
+29,qwen3,0.0,1.0
+30,qwen3,0.0,1.0
 

src/investment/investment_draw_boxplot.py

@@ -4,11 +4,15 @@ import matplotlib.pyplot as plt
 
 # Custom color palette
 color_palette = {
-    'random' : '#333333',  # Black
-    'gpt-4.5-preview-2025-02-27': '#7abaff',  # Blue
-    'llama3': '#32a68c',  # Green
-    'mistral-small': '#ff6941',  # Orange
-    'deepseek-r1': '#5862ed'  # Indigo
+    'random': '#333333',  # Black
+    'gpt-4.5-preview-2025-02-27': '#7abaff',  # BlueEscape
+    'llama3': '#32a68c',  # GreenFuture
+    'llama3.3:latest': '#4b9f7d',  # GreenLlama3.3
+    'mistral-small': '#ff6941',  # WarmOrange
+    'mixtral:8x7b': '#f1a61a',  # YellowMixtral
+    'deepseek-r1': '#5862ed',  # InclusiveIndigo
+    'deepseek-r1:7b': '#9a7bff',  # PurpleDeepseek-r1:7b
+    'qwen3': '#000000'
 }
 
 # Load CSV file

src/investment/investment_draw_violin.py

@@ -11,7 +11,8 @@ color_palette = {
     'mistral-small': '#ff6941',  # WarmOrange
     'mixtral:8x7b': '#f1a61a',  # YellowMixtral
     'deepseek-r1': '#5862ed',  # InclusiveIndigo
-    'deepseek-r1:7b': '#9a7bff'  # PurpleDeepseek-r1:7b
+    'deepseek-r1:7b': '#9a7bff',  # PurpleDeepseek-r1:7b
+    'qwen3': '#000000'
 }
 
 # Specify the order of models for the x-axis
@@ -20,7 +21,8 @@ model_order = [
     'gpt-4.5-preview-2025-02-27',
     'llama3', 'llama3.3:latest',  # Place llama3 and llama3.3:latest together
     'mistral-small', 'mixtral:8x7b',  # Bring mistral-small and mixtral:8x7b closer
-    'deepseek-r1', 'deepseek-r1:7b'
+    'deepseek-r1', 'deepseek-r1:7b',
+    'qwen3'
 ]
 
 

src/investment/investment_experiments.py

@@ -3,7 +3,7 @@ import csv
 from investment import Investment
 
 # Define models, temperature, and iterations
-models = ["deepseek-r1:7b"]  # "gpt-4.5-preview-2025-02-27", "optimal", "random", "llama3", "mistral-small", "deepseek-r1", "mixtral:8x7b", "llama3.3:latest",
+models = ["qwen3"]  # "gpt-4.5-preview-2025-02-27", "optimal", "random", "llama3", "mistral-small", "deepseek-r1", "mixtral:8x7b", "llama3.3:latest",
 temperature = 0.0
 iterations = 30
 output_file = "../../data/investment/investment.csv"