-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathexp_configs.py
122 lines (107 loc) · 3.13 KB
/
exp_configs.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
EXP_GROUPS = {}
# define the insights experiments
EXP_GROUPS["insights_wo_skills"] = []
for challenge in ["mid"]:
EXP_GROUPS["insights_wo_skills"] += [
{
"challenge": "mid",
"model": "gpt-4o",
"eval_mode": "insights",
"with_skills": 0,
}
]
EXP_GROUPS["insights_wo_skills_sai_latest"] = []
for challenge in ["toy"]:
EXP_GROUPS["insights_wo_skills_sai_latest"] += [
{
"challenge": "toy",
"model": "gpt-4o",
"eval_mode": "insights",
"with_skills": 0,
}
]
EXP_GROUPS["insights_w_skills_sai_latest"] = []
for challenge in ["toy"]:
EXP_GROUPS["insights_w_skills_sai_latest"] += [
{
"challenge": "toy",
"model": "gpt-4o",
"eval_mode": "insights",
"with_skills": 1,
}
]
EXP_GROUPS["insights_w_skills"] = []
for challenge in ["mid"]:
EXP_GROUPS["insights_w_skills"] += [
{
"challenge": "mid",
"model": "gpt-4o",
"eval_mode": "insights",
"with_skills": 1,
}
]
EXP_GROUPS["insights_w_skills_only"] = []
for challenge in ["toy"]:
EXP_GROUPS["insights_w_skills_only"] += [
{
"challenge": "toy",
"model": "gpt-4o",
"eval_mode": "insights_only",
"with_skills": 1,
}
]
EXP_GROUPS["insights_w_skills_only_planning"] = []
for challenge in ["toy"]:
EXP_GROUPS["insights_w_skills_only_planning"] += [
{
"challenge": "toy",
"model": "gpt-4o",
"eval_mode": "insights_only",
"with_skills": 1,
}
]
EXP_GROUPS["insights_wo_skills_only"] = []
for challenge in ["toy"]:
EXP_GROUPS["insights_wo_skills_only"] += [
{
"challenge": "toy",
"model": "gpt-4o",
"eval_mode": "insights_only",
"with_skills": 0,
}
]
# define the insights experiments
EXP_GROUPS["insights_prompts"] = []
for challenge in ["toy"]:
for prompt_strategy in ["basic", "advanced"]:
EXP_GROUPS["insights_prompts"] += [
{
"challenge": "toy",
"model": "gpt-4o-mini",
"eval_mode": "insights",
"with_skills": 1,
"prompt_strategy": prompt_strategy,
}
]
# define the skills experiments
EXP_GROUPS["skills"] = []
for challenge in ["full"]:
EXP_GROUPS["skills"] += [
{"challenge": "full", "model": "gpt-4o-mini", "eval_mode": "skills"}
]
# define the models experiments
# Check for duplicates in each experiment group
for group_name, experiments in EXP_GROUPS.items():
seen_configs = set()
duplicates = []
for exp in experiments:
# Convert dict to a hashable format (frozen set of items)
exp_tuple = frozenset(exp.items())
if exp_tuple in seen_configs:
duplicates.append(dict(exp_tuple))
else:
seen_configs.add(exp_tuple)
if duplicates:
raise ValueError(
f"Duplicate configurations found in {group_name}: {duplicates}"
)