Upload 96 files

dataset/Test Generation/ComplexCodeEval-Java/3/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~57),subset_1(59~85),subset_2(94~128)
+StarCoder2-15b,32.05,25.09,17.77
+CodeLlama-7b,33.09,25.34,23.49
+CodeLlama-13b,30.49,24.62,21.73
+CodeLlama-34b,31.61,24.30,23.39
+DeepSeek-Coder-1.3b,30.93,25.69,20.56
+DeepSeek-Coder-6.7b,20.22,23.69,22.39
+DeepSeek-Coder-33b,34.15,27.25,20.25

dataset/Test Generation/ComplexCodeEval-Java/3/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~377),subset_1(424~664),subset_2(681~952)   
+StarCoder2-15b,32.51,22.26,19.93
+CodeLlama-7b,33.77,22.98,25.15
+CodeLlama-13b,30.70,23.83,21.73
+CodeLlama-34b,31.85,23.58,24.12
+DeepSeek-Coder-1.3b,31.14,24.98,19.11
+DeepSeek-Coder-6.7b,20.65,21.81,23.54
+DeepSeek-Coder-33b,34.23,25.40,22.73

dataset/Test Generation/ComplexCodeEval-Java/3/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/3/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~37),subset_1(37~57),subset_2(57~128)
+StarCoder2-15b,31.93,32.36,22.45
+CodeLlama-7b,34.19,32.50,24.35
+CodeLlama-13b,30.66,30.53,23.52
+CodeLlama-34b,31.38,32.19,23.85
+DeepSeek-Coder-1.3b,29.87,32.10,23.90
+DeepSeek-Coder-6.7b,20.28,20.43,22.83
+DeepSeek-Coder-33b,34.69,33.71,24.80

dataset/Test Generation/ComplexCodeEval-Java/3/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~279),subset_1(279~377),subset_2(424~952)  
+StarCoder2-15b,30.79,34.29,21.70
+CodeLlama-7b,33.23,34.33,23.51
+CodeLlama-13b,29.66,31.77,23.32
+CodeLlama-34b,29.98,33.77,23.71
+DeepSeek-Coder-1.3b,28.72,33.62,23.56
+DeepSeek-Coder-6.7b,20.00,21.32,22.23
+DeepSeek-Coder-33b,33.30,35.19,24.75

dataset/Test Generation/ComplexCodeEval-Java/3/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/4/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~47),subset_1(51~74),subset_2(85~99),subset_3(109~128)
+StarCoder2-15b,33.11,25.25,20.71,20.50
+CodeLlama-7b,34.10,29.25,21.00,23.62
+CodeLlama-13b,31.07,27.33,21.22,23.23
+CodeLlama-34b,31.91,28.57,22.31,23.22
+DeepSeek-Coder-1.3b,30.85,29.82,22.11,22.76
+DeepSeek-Coder-6.7b,20.25,22.86,21.75,24.54
+DeepSeek-Coder-33b,35.08,29.94,21.64,21.86

dataset/Test Generation/ComplexCodeEval-Java/4/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~335),subset_1(346~544),subset_2(602~681),subset_3(793~952)
+StarCoder2-15b,29.65,32.16,21.97,22.42
+CodeLlama-7b,32.59,32.34,21.12,23.90
+CodeLlama-13b,29.19,30.96,21.81,21.52
+CodeLlama-34b,30.05,31.90,21.21,26.06
+DeepSeek-Coder-1.3b,29.05,32.14,23.49,18.65
+DeepSeek-Coder-6.7b,21.59,19.46,22.72,23.39
+DeepSeek-Coder-33b,32.06,34.73,23.98,20.64

dataset/Test Generation/ComplexCodeEval-Java/4/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/4/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~32),subset_1(32~44),subset_2(44~74),subset_3(74~128)
+StarCoder2-15b,32.69,31.06,30.98,21.04
+CodeLlama-7b,35.41,33.18,31.16,21.78
+CodeLlama-13b,30.73,30.46,30.14,21.72
+CodeLlama-34b,33.20,30.11,30.81,22.53
+DeepSeek-Coder-1.3b,28.44,31.39,32.32,22.39
+DeepSeek-Coder-6.7b,20.08,22.14,19.90,22.57
+DeepSeek-Coder-33b,32.05,36.05,34.47,21.83

dataset/Test Generation/ComplexCodeEval-Java/4/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~234),subset_1(241~349),subset_2(354~533),subset_3(533~952)
+StarCoder2-15b,33.94,27.49,33.30,21.04
+CodeLlama-7b,34.71,31.43,33.62,21.78
+CodeLlama-13b,30.54,28.56,32.23,21.72
+CodeLlama-34b,32.93,27.21,33.99,22.53
+DeepSeek-Coder-1.3b,27.61,30.73,33.81,22.39
+DeepSeek-Coder-6.7b,19.56,23.28,19.28,22.57
+DeepSeek-Coder-33b,32.95,31.98,37.64,21.83

dataset/Test Generation/ComplexCodeEval-Java/4/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/5/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~43),subset_1(44~63),subset_2(65~85),subset_3(94~99),subset_4(109~128)
+StarCoder2-15b,31.28,33.86,24.02,15.04,20.50
+CodeLlama-7b,33.31,32.86,24.18,23.35,23.62
+CodeLlama-13b,30.06,31.34,24.00,20.23,23.23
+CodeLlama-34b,30.25,34.21,23.72,23.56,23.22
+DeepSeek-Coder-1.3b,28.75,35.56,24.86,18.36,22.76
+DeepSeek-Coder-6.7b,21.24,18.71,23.31,20.24,24.54
+DeepSeek-Coder-33b,33.10,36.17,26.63,18.64,21.86

dataset/Test Generation/ComplexCodeEval-Java/5/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~209),subset_1(211~300),subset_2(300~365),subset_3(366~544),subset_4(602~952)
+StarCoder2-15b,30.01,36.39,21.63,18.67,22.42
+CodeLlama-7b,33.01,34.76,22.99,25.34,23.90
+CodeLlama-13b,29.56,32.30,23.94,21.82,21.52
+CodeLlama-34b,29.84,35.10,22.95,22.07,26.06
+DeepSeek-Coder-1.3b,28.67,35.08,24.86,21.22,18.65
+DeepSeek-Coder-6.7b,20.97,20.27,21.39,25.76,23.39
+DeepSeek-Coder-33b,32.12,37.38,25.53,26.03,20.64

dataset/Test Generation/ComplexCodeEval-Java/5/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/5/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models,subset_0(22~29),subset_1(30~39),subset_2(41~47),subset_3(47~85),subset_4(85~128)
+StarCoder2-15b,33.66,29.03,36.27,26.50,19.25
+CodeLlama-7b,35.78,29.99,36.10,27.81,22.25
+CodeLlama-13b,30.86,27.90,34.93,26.51,21.12
+CodeLlama-34b,33.25,27.24,35.53,27.23,22.56
+DeepSeek-Coder-1.3b,28.50,29.12,34.01,30.23,21.31
+DeepSeek-Coder-6.7b,20.00,21.47,19.97,22.38,22.03
+DeepSeek-Coder-33b,32.10,33.62,39.30,29.49,20.99

dataset/Test Generation/ComplexCodeEval-Java/5/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~209),subset_1(211~300),subset_2(300~365),subset_3(366~544),subset_4(602~952)
+StarCoder2-15b,33.66,27.58,36.39,24.99,22.10
+CodeLlama-7b,35.78,31.14,34.89,28.17,21.95
+CodeLlama-13b,30.86,28.94,33.44,26.35,21.72
+CodeLlama-34b,33.25,26.90,37.47,25.52,22.66
+DeepSeek-Coder-1.3b,28.50,29.56,34.82,28.26,22.04
+DeepSeek-Coder-6.7b,20.00,21.50,22.60,18.84,22.92
+DeepSeek-Coder-33b,32.10,33.27,38.51,28.64,22.98

dataset/Test Generation/ComplexCodeEval-Java/5/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/6/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~39),subset_1(41~57),subset_2(59~74),subset_3(85~85),subset_4(94~109),subset_5(127~128)
+StarCoder2-15b,31.34,33.05,27.40,23.55,14.39,24.52
+CodeLlama-7b,32.88,33.37,33.62,19.82,24.73,20.99
+CodeLlama-13b,29.38,32.07,28.99,21.71,20.76,23.66
+CodeLlama-34b,30.25,33.56,28.22,21.69,22.25,25.67
+DeepSeek-Coder-1.3b,28.81,33.95,28.24,23.99,18.90,23.89
+DeepSeek-Coder-6.7b,20.73,19.47,25.46,22.51,21.18,24.81
+DeepSeek-Coder-33b,32.86,35.99,33.40,23.15,21.23,18.29

dataset/Test Generation/ComplexCodeEval-Java/6/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~273),subset_1(276~377),subset_2(424~544),subset_3(602~664),subset_4(681~793),subset_5(935~952)
+StarCoder2-15b,31.20,33.64,21.07,23.55,13.83,23.59
+CodeLlama-7b,33.61,33.91,25.90,19.82,28.42,23.19
+CodeLlama-13b,30.26,31.07,25.79,21.71,20.41,22.53
+CodeLlama-34b,30.96,32.61,25.32,21.69,20.93,26.04
+DeepSeek-Coder-1.3b,29.47,32.58,25.89,23.99,16.45,20.71
+DeepSeek-Coder-6.7b,20.40,20.86,21.17,22.51,24.54,22.94
+DeepSeek-Coder-33b,34.42,34.07,27.48,23.15,28.31,19.38

dataset/Test Generation/ComplexCodeEval-Java/6/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/6/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~28),subset_1(29~37),subset_2(37~44),subset_3(44~57),subset_4(59~85),subset_5(85~128)
+StarCoder2-15b,33.42,30.44,32.66,31.66,26.63,18.06
+CodeLlama-7b,35.78,32.60,33.45,30.50,27.62,21.68
+CodeLlama-13b,30.27,31.05,30.77,29.86,27.03,20.04
+CodeLlama-34b,32.97,29.78,32.43,31.26,25.44,22.48
+DeepSeek-Coder-1.3b,28.15,31.59,30.67,33.30,27.95,19.59
+DeepSeek-Coder-6.7b,18.13,22.43,23.05,17.26,24.35,22.05
+DeepSeek-Coder-33b,31.33,38.05,33.66,33.55,29.14,20.10

dataset/Test Generation/ComplexCodeEval-Java/6/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~205),subset_1(205~279),subset_2(279~354),subset_3(358~424),subset_4(486~606),subset_5(606~952)
+StarCoder2-15b,31.09,30.49,29.57,38.64,21.92,21.08
+CodeLlama-7b,32.92,33.55,31.68,36.57,25.33,21.44
+CodeLlama-13b,29.05,30.27,29.62,33.49,25.47,21.11
+CodeLlama-34b,31.60,28.37,29.74,37.70,23.95,22.95
+DeepSeek-Coder-1.3b,26.90,30.55,29.58,37.45,25.08,21.63
+DeepSeek-Coder-6.7b,20.31,19.69,24.35,18.45,21.75,22.60
+DeepSeek-Coder-33b,31.32,35.27,30.36,39.57,26.67,22.65

dataset/Test Generation/ComplexCodeEval-Java/6/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/7/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~205),subset_1(205~279),subset_2(279~354),subset_3(358~424),subset_4(486~606),subset_5(606~952)
+StarCoder2-15b,31.29,33.67,27.94,24.39,20.96,13.83,24.52
+CodeLlama-7b,33.59,33.65,25.37,34.67,20.62,28.42,20.99
+CodeLlama-13b,29.62,31.82,25.93,30.93,21.50,20.41,23.66
+CodeLlama-34b,30.52,33.61,24.26,29.43,22.09,20.93,25.67
+DeepSeek-Coder-1.3b,28.96,33.58,27.15,28.66,22.93,16.45,23.89
+DeepSeek-Coder-6.7b,20.37,20.67,19.51,25.49,21.52,24.54,24.81
+DeepSeek-Coder-33b,33.64,34.81,32.11,34.68,21.33,28.31,18.29

dataset/Test Generation/ComplexCodeEval-Java/7/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~255),subset_1(259~367),subset_2(373~486),subset_3(490~602),subset_4(606~681),subset_5(793~793),subset_6(935~952)
+StarCoder2-15b,31.18,33.11,30.92,20.98,21.56,16.57,23.59
+CodeLlama-7b,33.77,33.36,29.04,28.25,20.22,27.45,23.19
+CodeLlama-13b,30.68,30.95,26.74,25.39,21.88,16.48,22.53
+CodeLlama-34b,30.98,33.69,21.87,25.83,20.56,26.15,26.04
+DeepSeek-Coder-1.3b,29.52,32.18,31.15,24.27,24.27,8.33,20.71
+DeepSeek-Coder-6.7b,19.88,22.09,13.21,23.21,22.51,25.63,22.94
+DeepSeek-Coder-33b,34.50,33.72,34.46,26.08,23.97,26.97,19.38

dataset/Test Generation/ComplexCodeEval-Java/7/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/7/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~26),subset_1(27~36),subset_2(36~42),subset_3(42~45),subset_4(46~70),subset_5(70~85),subset_6(85~128)
+StarCoder2-15b,34.01,28.36,31.34,37.87,28.95,23.71,18.04
+CodeLlama-7b,36.81,33.41,30.02,35.05,29.66,23.48,23.57
+CodeLlama-13b,30.91,32.25,27.06,34.66,26.05,25.24,21.20
+CodeLlama-34b,32.95,30.63,26.76,40.14,26.25,23.39,23.65
+DeepSeek-Coder-1.3b,29.10,31.76,26.04,35.73,30.30,27.63,19.63
+DeepSeek-Coder-6.7b,18.24,21.67,22.89,20.17,19.95,22.88,22.57
+DeepSeek-Coder-33b,30.65,38.49,30.31,40.46,29.93,27.16,20.20

dataset/Test Generation/ComplexCodeEval-Java/7/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~183),subset_1(187~259),subset_2(273~312),subset_3(330~358),subset_4(358~508),subset_5(530~606),subset_6(606~952)
+StarCoder2-15b,30.09,32.14,28.25,38.02,31.59,21.33,20.87
+CodeLlama-7b,31.89,35.35,30.66,36.52,30.35,25.93,21.53
+CodeLlama-13b,27.75,32.32,27.99,35.03,28.56,25.61,20.33
+CodeLlama-34b,30.76,30.96,27.46,38.94,28.51,24.47,22.80
+DeepSeek-Coder-1.3b,27.06,31.64,27.11,38.68,29.50,25.86,20.50
+DeepSeek-Coder-6.7b,19.38,20.49,24.06,20.51,18.16,23.44,22.34
+DeepSeek-Coder-33b,30.38,38.09,26.98,41.58,32.81,25.19,22.22

dataset/Test Generation/ComplexCodeEval-Java/7/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/8/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~35),subset_1(36~47),subset_2(51~59),subset_3(63~74),subset_4(85~85),subset_5(94~99),subset_6(109~109),subset_7(127~128)
+StarCoder2-15b,31.16,34.87,25.15,25.38,23.55,15.04,12.46,24.52
+CodeLlama-7b,35.27,33.04,26.75,32.12,19.82,23.35,28.90,20.99
+CodeLlama-13b,31.56,30.63,26.74,27.99,21.71,20.23,22.37,23.66
+CodeLlama-34b,31.88,31.95,28.72,28.39,21.69,23.56,18.32,25.67
+DeepSeek-Coder-1.3b,30.50,31.16,32.20,27.09,23.99,18.36,20.52,23.89
+DeepSeek-Coder-6.7b,20.15,20.35,21.16,24.80,22.51,20.24,24.00,24.81
+DeepSeek-Coder-33b,34.52,35.59,27.48,32.75,23.15,18.64,28.98,18.29

dataset/Test Generation/ComplexCodeEval-Java/8/EI/token_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~234),subset_1(241~335),subset_2(346~424),subset_3(486~544),subset_4(602~608),subset_5(664~681),subset_6(793~793),subset_7(935~952)
+StarCoder2-15b,33.94,24.98,39.16,20.49,22.87,18.67,16.57,23.59
+CodeLlama-7b,34.71,30.29,36.29,25.76,19.97,25.34,27.45,23.19
+CodeLlama-13b,30.54,27.72,34.00,25.90,21.80,21.82,16.48,22.53
+CodeLlama-34b,32.93,26.91,36.18,24.77,20.97,22.07,26.15,26.04
+DeepSeek-Coder-1.3b,27.61,30.61,36.09,25.55,24.11,21.22,8.33,20.71
+DeepSeek-Coder-6.7b,19.56,23.79,18.57,20.93,21.89,25.76,25.63,22.94
+DeepSeek-Coder-33b,32.95,31.10,39.09,27.46,23.42,26.03,26.97,19.38

dataset/Test Generation/ComplexCodeEval-Java/8/EI/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Java/8/QS/line_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(22~26),subset_1(26~32),subset_2(33~38),subset_3(39~44),subset_4(44~51),subset_5(51~74),subset_6(85~85),subset_7(94~128)
+StarCoder2-15b,33.53,31.84,29.20,35.55,32.95,25.96,23.55,17.77
+CodeLlama-7b,37.14,34.38,28.18,36.84,32.22,29.76,19.82,23.49
+CodeLlama-13b,30.62,31.71,26.14,34.43,31.27,27.68,21.71,21.73
+CodeLlama-34b,33.27,33.66,24.49,36.26,30.75,28.87,21.69,23.39
+DeepSeek-Coder-1.3b,27.48,29.11,30.19,34.76,30.91,31.49,23.99,20.56
+DeepSeek-Coder-6.7b,16.65,22.56,22.65,21.89,18.83,21.96,22.51,22.39
+DeepSeek-Coder-33b,29.25,35.16,35.16,38.43,34.81,31.47,23.15,20.25

dataset/Test Generation/ComplexCodeEval-Java/8/QS/token_counts_QS.csv

@@ -0,0 +1,8 @@
+Models, subset_0(139~179),subset_1(181~241),subset_2(241~290),subset_3(300~358),subset_4(358~367),subset_5(373~533),subset_6(544~606),subset_7(606~952)
+StarCoder2-15b,30.41,35.66,25.70,31.79,38.08,27.93,21.87,19.46
+CodeLlama-7b,32.83,37.73,29.81,32.13,35.39,30.97,21.23,22.08
+CodeLlama-13b,27.04,36.15,26.33,30.35,34.57,27.66,23.15,20.30
+CodeLlama-34b,30.22,33.89,25.98,31.87,40.98,24.83,21.86,23.22
+DeepSeek-Coder-1.3b,26.76,30.70,29.82,30.10,38.03,28.88,25.47,19.08
+DeepSeek-Coder-6.7b,18.59,20.53,22.52,23.10,20.67,19.06,22.10,22.80
+DeepSeek-Coder-33b,29.98,38.19,30.10,33.34,39.18,34.03,22.20,21.20

dataset/Test Generation/ComplexCodeEval-Java/8/QS/tongji.py

@@ -0,0 +1,50 @@
+import json
+from collections import defaultdict
+import os
+
+def analyze_json_file(file_path):
+    # 读取JSON文件
+    with open(file_path, 'r',encoding="utf-8") as f:
+        data = json.load(f)
+    
+    # 初始化统计字典
+    line_diff_stats = defaultdict(list)
+    token_diff_stats = defaultdict(list)
+    
+    # 收集数据
+    for entry in data:
+        line_diff = entry['line_diff']
+        token_diff = entry['token_diff']
+        line = entry['line']
+        token = entry['token']
+        
+        line_diff_stats[line_diff].append(line)
+        token_diff_stats[token_diff].append(token)
+    
+    # 处理line_diff统计结果
+    print("Models: ", end="")
+    line_diff_keys = sorted(line_diff_stats.keys())
+    line_subsets = []
+    for diff in line_diff_keys:
+        lines = line_diff_stats[diff]
+        min_line = min(lines)
+        max_line = max(lines)
+        line_subsets.append(f"subset_{diff}({min_line}~{max_line})")
+    print(",".join(line_subsets))
+    
+    # 处理token_diff统计结果
+    print("Models: ", end="")
+    token_diff_keys = sorted(token_diff_stats.keys())
+    token_subsets = []
+    for diff in token_diff_keys:
+        tokens = token_diff_stats[diff]
+        min_token = min(tokens)
+        max_token = max(tokens)
+        token_subsets.append(f"subset_{diff}({min_token}~{max_token})")
+    print(",".join(token_subsets))
+
+# 使用示例
+if __name__ == "__main__":
+    # 假设JSON文件名为data.json
+    file_path = "EI.json" if os.path.exists("EI.json") else "QS.json"
+    analyze_json_file(file_path)

dataset/Test Generation/ComplexCodeEval-Python/3/EI/line_counts_EI.csv

@@ -0,0 +1,8 @@
+Models, subset_0(15~187), subset_1,subset_2(749~749)
+StarCoder2-15b,24.55,0.00,20.70
+CodeLlama-7b,26.44,0.00,27.18
+CodeLlama-13b,26.46,0.00,31.82
+CodeLlama-34b,27.20,0.00,24.58
+DeepSeek-Coder-1.3b,25.85,0.00,21.28
+DeepSeek-Coder-6.7b,24.76,0.00,20.74
+DeepSeek-Coder-33b,29.22,0.00,25.46