-
Notifications
You must be signed in to change notification settings - Fork 34
/
vis_util.py
128 lines (111 loc) · 5 KB
/
vis_util.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
123
124
125
126
127
128
# Copyright 2022 The FastEstimator Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
import json
import os
from collections import defaultdict
from typing import Any, Optional, Sequence, Union
from natsort import humansorted
from fastestimator.search.search import Search
from fastestimator.summary.summary import ValWithError
from fastestimator.util.base_util import to_set
def _load_search_file(path: str) -> Search:
path = os.path.abspath(os.path.normpath(path))
if not os.path.exists(path):
raise ValueError(f"No file found at location: {path}")
with open(path, 'r') as file:
state = json.load(file)
search = Search.__new__(Search)
search.name = 'Search'
search.best_mode = None
search.optimize_field = None
search._initialize_state()
search.__dict__.update(state)
return search
class SearchData:
def __init__(self, search: Search, ignore_keys: Union[None, str, Sequence[str]] = None):
self.params = []
self.results = []
self.data = defaultdict(list)
self.categorical_maps = {}
self.ignored_params = False # Did you ignore a param other than search_idx which had more than 1 value?
search = search.get_search_summary()
if not search:
return
example_item = search[0]
self.params = set(example_item['param'].keys())
self.results = set(example_item['result'].keys())
ignore_keys = to_set(ignore_keys) | {'search_idx'}
for key in ignore_keys:
self.params.discard(key)
self.results.discard(key)
# Check if any results left to use
assert len(self.results) != 0, f"No results found after ignoring keys: {ignore_keys}"
# Keep a sample parameter value to catch boring parameters
param_samples = {}
for elem in search:
pars = elem['param']
for k, v in pars.items():
if k in ignore_keys:
if k != 'search_idx' and v != param_samples.setdefault(k, v):
# The ignored key had more than 1 value
self.ignored_params = True
continue
if k not in self.params:
raise ValueError("Inconsistent parameter list detected")
v = self._parse_value(v)
param_samples.setdefault(k, v)
self.data[k].append(v)
res = elem['result']
for k, v in res.items():
if k in ignore_keys:
continue
if k not in self.results:
raise ValueError("Inconsistent result list detected")
self.data[k].append(self._parse_value(v))
# Remove any parameters which have only 1 value since they are boring to visualize
for param in list(self.params): # Copy to a list since it may be modified during iteration
for val in self.data[param]:
if val != param_samples[param]:
break
else:
self.params.discard(param)
self.data.pop(param)
# Handle categorical data
for key, values in self.data.items():
if all([isinstance(value, (int, float)) for value in values]):
continue # Numeric value
# Else categorical
categories = humansorted(set(values), reverse=True)
self.categorical_maps[key] = {cat: i for i, cat in enumerate(categories)}
self.data[key] = [self.categorical_maps[key][val] for val in values]
self.params = humansorted(self.params)
self.results = humansorted(self.results)
self.inverse_maps = {key: {v2: k2 for k2, v2 in val.items()} for key, val in self.categorical_maps.items()}
def to_category(self, key: str, val: Any) -> Optional[str]:
m = self.inverse_maps.get(key, {})
if not m:
return val if val is None else str(val)
return m[val]
@staticmethod
def _parse_value(value: Any) -> Union[int, float, str, None]:
if isinstance(value, (list, tuple)) and len(value) == 1:
value = value[0]
if hasattr(value, 'item') and hasattr(value, 'size') and value.size == 1:
value = value.item()
if isinstance(value, ValWithError):
value = value.y
if not isinstance(value, (int, float, str, type(None))):
value = str(value)
return value