forked from iarai/weather4cast
-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.py
187 lines (165 loc) · 8.92 KB
/
config.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
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
# Author: Pedro Herruzo
# Copyright 2021 Institute of Advanced Research in Artificial Intelligence (IARAI) GmbH.
# IARAI licenses this file to You 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 numpy as np
import os
def prepare_crop(regions, region_id):
""" this function prepares the expected parameters to crop images per region
e.g., to crop latitudes to the region of interest
"""
x, y = regions[region_id]['up_left']
crop = {'x_start': x, 'y_start': y, 'size': regions[region_id]['size']}
return crop
def n_extra_vars(string_vars):
""" computes how many extra variables will be used """
if string_vars=='':
len_extra = 0
else:
len_extra = len(string_vars.split('-'))
if 'l' in string_vars:
len_extra += 1 # 'l' loads both lat/lon, so 2 vars (not 1)
return len_extra
def get_prod_name(product):
""" get the folder name for each product. Note that only the folder containing ASII
have a slightly different name
"""
if product=='ASII':
product = 'ASII-TF'
return product
def get_params(region_id='R1',
data_path=os.path.join(os.getcwd(), '../data'),
splits_path=os.path.join(os.getcwd()),
static_data_path=os.path.join(os.getcwd(), '../data/static'),
size=256,
collapse_time=False):
""" Set paths & parameters to load/transform/save data and models.
Args:
region_id (str, optional): Region to load data from]. Defaults to 'R1'.
data_path (str, optional): path to the parent folder containing folders
for the core competition (*/w4c-core-stage-1) and/or
transfer learning comptition (*/w4c-transfer-learning-stage-1').
Defaults to 'data'.
splits_path (str, optional): Path to the folder containing the csv and json files defining
the data splits.
Defaults to 'utils'.
static_data_path (str, optional): Path to the folder containing the static channels.
Defaults to 'data/static'.
size (int, optional): Size of the region. Default to 256.
Returns:
dict: Contains the params
"""
data_params = {}
model_params = {}
training_params = {}
optimization_params = {}
regions = {'R3': {'up_left': (935, 400), 'split': 'train', 'desc': 'South West\nEurope', 'size': size},
'R6': {'up_left': (1270, 250), 'split': 'test', 'desc': 'Central\nEurope', 'size': size},
'R2': {'up_left': (1550, 200), 'split': 'train', 'desc': 'Eastern\nEurope', 'size': size},
'R1': {'up_left': (1850, 760), 'split': 'train', 'desc': 'Nile Region', 'size': size},
'R5': {'up_left': (1300, 550), 'split': 'test', 'desc': 'South\nMediterranean', 'size': size},
'R4': {'up_left': (1020, 670), 'split': 'test', 'desc': 'Central\nMaghreb', 'size': size},
'R7': {'up_left': (1700, 470), 'split': 'train', 'desc': 'Bosphorus', 'size': size},
'R8': {'up_left': (750, 670), 'split': 'train', 'desc': 'East\nMaghreb', 'size': size},
'R9': {'up_left': (450, 760), 'split': 'test', 'desc': 'Canarian Islands', 'size': size},
'R10': {'up_left': (250, 500), 'split': 'test', 'desc': 'Azores Islands', 'size': size},
'R11': {'up_left': (1000, 130), 'split': 'test', 'desc': 'North West\nEurope','size': size}
}
print(f'Using data for region {region_id} | size: {size} | {regions[region_id]["desc"]}')
# ------------
# 1. Files to load
# ------------
if region_id in ['R1', 'R2', 'R3', 'R7', 'R8']:
track = 'core-w4c'
else:
track = 'transfer-learning-w4c'
data_params['data_path'] = os.path.join(data_path, track, region_id)
data_params['static_paths'] = {}
data_params['static_paths']['l'] = os.path.join(static_data_path, 'Navigation_of_S_NWC_CT_MSG4_Europe-VISIR_20201106T120000Z.nc')
data_params['static_paths']['e'] = os.path.join(static_data_path, 'S_NWC_TOPO_MSG4_+000.0_Europe-VISIR.raw')
data_params['train_splits'] = os.path.join(splits_path, 'splits.csv')
data_params['test_splits'] = os.path.join(splits_path, 'test_split.json')
data_params['black_list_path'] = os.path.join(splits_path, 'blacklist.json')
# ------------
# 2. Data params
# ------------
data_params['collapse_time'] = collapse_time
data_params['extra_data'] = 'l-e' # use '' to not use static features
data_params['target_vars'] = ['temperature', 'crr_intensity', 'asii_turb_trop_prob', 'cma']
data_params['products'] = {'CTTH': ['temperature'],
'CRR': ['crr_intensity'],
'ASII': ['asii_turb_trop_prob'],
'CMA': ['cma']}
data_params['weigths'] = {'temperature': .25,
'crr_intensity': .25,
'asii_turb_trop_prob': .25,
'cma': .25} # to use by the metric
data_params['depth'] = len(data_params['target_vars']) + n_extra_vars(data_params['extra_data']) + 1 # lead time is added
data_params['spatial_dim'] = (size, size)
data_params['crop_static'] = prepare_crop(regions, region_id)
data_params['crop_in'] = None
data_params['crop_out'] = None
data_params['train_region_id'] = region_id+'_mse'*1 # this is actually used by the model, not the data ??????
data_params['region_id'] = region_id
data_params['len_seq_in'] = 4 # time-bins of 15 minutes
data_params['len_seq_out'] = 1 # time-bins
data_params['bins_to_predict'] = 8*4 # hours x (time-bins per hour =4) # not used
data_params['day_bins'] = 96
data_params['seq_mode'] = 'sliding_window' # not used
data_params['width'] = 256 # not used
data_params['height'] = 256 # not used
# preprocessing:
# a. fill_value: value to replace NaNs (currently temperature is the one that has more)
# b. max_value: maximum value of the variable when it's saved on disk as integer
# c. scale_factor: netCDF automatically uses this value to re-scale the value
# d. add_offset: netCDF automatically uses this value to shift a variable
#
# c. and d. together mean that once loaded, the data is in the scale [add_offset, max_value*scale_factor + add_offset]
# Hence, to normalize the data between [0, 1] we must use:
# data = (data-add_offset)/(max_value*scale_factor - add_offset)
preprocess = {'cma': {'fill_value': 0, 'max_value': 1, 'add_offset': 0, 'scale_factor': 1},
'temperature': {'fill_value': 0, 'max_value': 35000, 'add_offset': 130, 'scale_factor': np.float32(0.01)},
'crr_intensity': {'fill_value': 0, 'max_value': 500, 'add_offset': 0, 'scale_factor': np.float32(0.1)},
'asii_turb_trop_prob': {'fill_value': 0, 'max_value': 100, 'add_offset': 0, 'scale_factor': 1}}
preprocess_tgt = {'cma': {'fill_value': np.nan, 'max_value': 1, 'add_offset': 0, 'scale_factor': 1},
'temperature': {'fill_value': np.nan, 'max_value': 35000, 'add_offset': 130, 'scale_factor': np.float32(0.01)},
'crr_intensity': {'fill_value': np.nan, 'max_value': 500, 'add_offset': 0, 'scale_factor': np.float32(0.1)},
'asii_turb_trop_prob': {'fill_value': np.nan, 'max_value': 100, 'add_offset': 0, 'scale_factor': 1}}
data_params['preprocess'] = {'source': preprocess, 'target': preprocess_tgt}
# ------------
# 3. Model params
# ------------
if data_params['collapse_time']:
model_params['in_channels'] = data_params['depth'] * data_params['len_seq_in']
else:
model_params['in_channels'] = data_params['depth']
model_params['n_classes'] = len(data_params['target_vars'])
model_params['depth'] = 5
model_params['wf'] = 6
model_params['padding'] = True
model_params['batch_norm'] = False
model_params['up_mode'] = 'upconv'
# ------------
# 4. Training params
# ------------
training_params['batch_size'] = 64
training_params['n_workers'] = 8
params = {
'data_params': data_params,
'model_params': model_params,
'training_params': training_params,
'optimization_params': optimization_params,
}
return params
if __name__ == '__main__':
# this is only executed when the module is run directly.
print(get_params())