|
1
|
|
|
# Copyright 2014 Diamond Light Source Ltd. |
|
2
|
|
|
# |
|
3
|
|
|
# Licensed under the Apache License, Version 2.0 (the "License"); |
|
4
|
|
|
# you may not use this file except in compliance with the License. |
|
5
|
|
|
# You may obtain a copy of the License at |
|
6
|
|
|
# |
|
7
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0 |
|
8
|
|
|
# |
|
9
|
|
|
# Unless required by applicable law or agreed to in writing, software |
|
10
|
|
|
# distributed under the License is distributed on an "AS IS" BASIS, |
|
11
|
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
|
12
|
|
|
# See the License for the specific language governing permissions and |
|
13
|
|
|
# limitations under the License. |
|
14
|
|
|
|
|
15
|
|
|
""" |
|
16
|
|
|
.. module:: comparison |
|
17
|
|
|
:platform: Unix |
|
18
|
|
|
:synopsis: A plugin to compare two datasets, given as input datasets, and print the RMSD between the two. |
|
19
|
|
|
The data is unchanged. |
|
20
|
|
|
|
|
21
|
|
|
.. moduleauthor:: Jacob Williamson <[email protected]> |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
|
|
from savu.plugins.utils import register_plugin |
|
25
|
|
|
from savu.plugins.plugin import Plugin |
|
26
|
|
|
from savu.plugins.driver.cpu_plugin import CpuPlugin |
|
27
|
|
|
from savu.core.iterate_plugin_group_utils import enable_iterative_loop, \ |
|
28
|
|
|
check_if_end_plugin_in_iterate_group, setup_extra_plugin_data_padding |
|
29
|
|
|
|
|
30
|
|
|
import os |
|
31
|
|
|
import h5py as h5 |
|
32
|
|
|
|
|
33
|
|
|
# This decorator is required for the configurator to recognise the plugin |
|
34
|
|
|
@register_plugin |
|
35
|
|
|
class GatherStats(Plugin, CpuPlugin): |
|
36
|
|
|
|
|
37
|
|
|
def __init__(self): |
|
38
|
|
|
super(GatherStats, self).__init__("GatherStats") |
|
39
|
|
|
|
|
40
|
|
|
def nInput_datasets(self): |
|
41
|
|
|
return 1 |
|
42
|
|
|
|
|
43
|
|
|
|
|
44
|
|
|
def nOutput_datasets(self): |
|
45
|
|
|
return 0 |
|
46
|
|
|
|
|
47
|
|
|
def nClone_datasets(self): |
|
48
|
|
|
if check_if_end_plugin_in_iterate_group(self.exp): |
|
49
|
|
|
return 1 |
|
50
|
|
|
else: |
|
51
|
|
|
return 0 |
|
52
|
|
|
|
|
53
|
|
|
|
|
54
|
|
|
@enable_iterative_loop |
|
55
|
|
|
def setup(self): |
|
56
|
|
|
|
|
57
|
|
|
in_dataset, out_dataset = self.get_datasets() |
|
58
|
|
|
self.stats_obj.calc_stats = False |
|
59
|
|
|
self.stats_obj.set_stats_key(["max", "min", "mean", "mean_std_dev", "median_std_dev", "zeros", "zeros%", |
|
60
|
|
|
"range_used"]) |
|
61
|
|
|
in_pData, out_pData = self.get_plugin_datasets() |
|
62
|
|
|
|
|
63
|
|
|
# Each plugin dataset must call this method and define the data access |
|
64
|
|
|
# pattern and number of frames required. |
|
65
|
|
|
for i in range(len(in_pData)): |
|
66
|
|
|
in_pData[i].plugin_data_setup(self.parameters['pattern'], 'single') |
|
67
|
|
|
|
|
68
|
|
|
# All dataset information can be accessed via the Data and PluginData |
|
69
|
|
|
# instances |
|
70
|
|
|
|
|
71
|
|
|
|
|
72
|
|
|
def pre_process(self): |
|
73
|
|
|
# This method is called once before any processing has begun. |
|
74
|
|
|
# Access parameters from the doc string in the parameters dictionary |
|
75
|
|
|
# e.g. self.parameters['example'] |
|
76
|
|
|
in_datasets = self.get_in_datasets() |
|
77
|
|
|
|
|
78
|
|
|
def process_frames(self, data): |
|
79
|
|
|
self.stats_obj.set_slice_stats(data, pad=False) |
|
80
|
|
|
return None |
|
81
|
|
|
|
|
82
|
|
|
def post_process(self): |
|
83
|
|
|
slice_stats = self.stats_obj.stats |
|
84
|
|
|
comm = self.get_communicator() |
|
85
|
|
|
combined_stats = self.stats_obj._combine_mpi_stats(slice_stats, comm=comm) |
|
86
|
|
|
|
|
87
|
|
|
volume_stats = self.stats_obj.calc_volume_stats(combined_stats) |
|
88
|
|
|
if self.exp.meta_data.get("pre_run"): |
|
89
|
|
|
|
|
90
|
|
|
self._generate_warnings(volume_stats) |
|
91
|
|
|
self.exp.meta_data.set("pre_run_stats", volume_stats) |
|
92
|
|
|
|
|
93
|
|
|
folder = self.exp.meta_data['out_path'] |
|
94
|
|
|
fname = self.exp.meta_data.get('datafile_name') + '_pre_run.nxs' |
|
95
|
|
|
filename = os.path.join(folder, fname) |
|
96
|
|
|
stats_array = self.stats_obj._dict_to_array(volume_stats) |
|
97
|
|
|
if comm.rank == 0: |
|
98
|
|
|
with h5.File(filename, "a") as h5file: |
|
99
|
|
|
fsplit = self.exp.meta_data["data_path"].split("/") |
|
100
|
|
|
fsplit[-1] = "" |
|
101
|
|
|
stats_path = "/".join(fsplit) |
|
102
|
|
|
stats_group = h5file.require_group(stats_path) |
|
103
|
|
|
dataset = stats_group.create_dataset("stats", shape=stats_array.shape, dtype=stats_array.dtype) |
|
104
|
|
|
dataset[::] = stats_array[::] |
|
105
|
|
|
dataset.attrs.create("stats_key", list(self.stats_obj.stats_key)) |
|
106
|
|
|
|
|
107
|
|
|
def _generate_warnings(self, volume_stats): |
|
108
|
|
|
warnings = [] |
|
109
|
|
|
if volume_stats["zeros%"] > 10: |
|
110
|
|
|
warnings.append(f"Percentage of data points that are 0s is {volume_stats['zeros%']}") |
|
111
|
|
|
if volume_stats["range_used"] < 2: |
|
112
|
|
|
warnings.append(f"Only {volume_stats['range_used']}% of the possible range of the datatype (\ |
|
113
|
|
|
{self.stats_obj.stats['dtype']}) has been used. The datatype used, {self.stats_obj.stats['dtype']} can go from \ |
|
114
|
|
|
{self.stats_obj.stats['possible_min']} to {self.stats_obj.stats['possible_max']}") |
|
115
|
|
|
self.exp.meta_data.set("warnings", warnings) |
|
116
|
|
|
|
DiamondLightSource /
Savu
