-
Notifications
You must be signed in to change notification settings - Fork 6
/
4-2.predict_pos.py
197 lines (180 loc) · 8 KB
/
4-2.predict_pos.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
188
189
190
191
192
193
194
195
196
197
import numpy as np
import argparse
import sys
import os
import shutil
import json
import time
import warnings
from random import sample
import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
from torch.autograd import Variable
from torch.optim.lr_scheduler import ExponentialLR
import matplotlib.pyplot as plt
from source.GCN_pos import SemiFullGN
from source.data_pos import collate_pool, get_data_loader, CIFData
from tqdm import tqdm
from source.models_cryslator import Generator
from source.GCN import GCN
def cal_cell_volume(cell):
batch_size = cell.shape[0]
result = []
for i in range(batch_size):
c = cell[i]
v = np.inner(np.cross(c[0,:],c[1,:]),c[2,:])
result.append(v)
return np.array(result)
def load_gcn(gcn_name):
checkpoint = torch.load(gcn_name)
x = checkpoint['model_args']
N_tr= x['N_tr']
N_val = x['N_val']
N_test = x['N_test']
atom_fea_len = x['atom_fea_len']
h_fea_len = x['h_fea_len']
n_conv = x['n_conv']
n_h = x['n_h']
orig_atom_fea_len = x['orig_atom_fea_len']
nbr_fea_len = x['nbr_fea_len']
model =GCN(orig_atom_fea_len,nbr_fea_len,atom_fea_len,n_conv,h_fea_len,n_h)
model.cuda()
model.load_state_dict(checkpoint['state_dict'])
model.eval()
return N_tr, N_val, N_test, orig_atom_fea_len, nbr_fea_len, model
def load_model(model_name,orig_atom_fea_len,nbr_fea_len):
checkpoint = torch.load(model_name)
x = checkpoint['model_args']
N_tr= x['N_tr']
N_val = x['N_val']
N_test = x['N_test']
atom_fea_len = x['atom_fea_len']
h_fea_len = x['h_fea_len']
n_conv = x['n_conv']
n_h = x['n_h']
orig_atom_fea_len = orig_atom_fea_len+3
nbr_fea_len = nbr_fea_len
model = SemiFullGN(orig_atom_fea_len,nbr_fea_len,atom_fea_len,n_conv,h_fea_len,n_h,n_feature=256)
model.cuda()
model.load_state_dict(checkpoint['state_dict'])
model.eval()
return model
def main():
#taken from sys.argv
model_folder = 'saved_models'
best_name = model_folder+'/'+'best_pos'
best_gcn = model_folder+'/'+'best_xmno'
N_tr, N_val, N_test, orig_atom_fea_len, nbr_fea_len, gcn = load_gcn(best_gcn)
model = load_model(best_name,orig_atom_fea_len, nbr_fea_len)
root_dir ='../Cryslator/data/jsons_xmno_rcut6/'
root_dir_pos ='../Cryslator/data/pos_xmno/'
#root_dir3 ='./data/feature_xmno/'
root_dir_cell ='../Cryslator/data/cell_xmno/'
max_num_nbr = 8
radius = 6
dmin = 0
step = 0.2
random_seed = 1234
batch_size = 4
N_tot = 28579 #full data
N_tr = int(N_tot*0.8)
N_val = int(N_tot*0.2)
N_test = N_tot - N_tr - N_val
train_idx = list(range(N_tr))
val_idx = list(range(N_tr,N_tr+N_val))
test_idx = list(range(N_tr+N_val,N_tr+N_val+N_test))
num_workers = 0
pin_memory = False
return_test = True
#var for model
train_csv = root_dir+'/'+'id_prop_train_all.csv'
val_csv = root_dir+'/'+'id_prop_val_all.csv'
test_csv = root_dir+'/'+'id_prop_test_all.csv'
train_dataset = CIFData(root_dir,root_dir_pos,root_dir_cell,train_csv,radius,dmin,step,random_seed)
val_dataset = CIFData(root_dir,root_dir_pos,root_dir_cell,val_csv,radius,dmin,step,random_seed)
test_dataset = CIFData(root_dir,root_dir_pos,root_dir_cell,test_csv,radius,dmin,step,random_seed)
collate_fn = collate_pool
train_loader = get_data_loader(train_dataset,collate_fn,batch_size,num_workers,pin_memory,False)
val_loader = get_data_loader(val_dataset,collate_fn,batch_size,num_workers,pin_memory,True)
test_loader= get_data_loader(test_dataset,collate_fn,batch_size,num_workers,pin_memory,True)
generator = Generator(4,4,4).cuda()
generator.load_state_dict(torch.load(model_folder+'/'+'best_G.pth'))
generator.eval()
# pos_save_folder = './pos_save_xmno_notchange'
pos_save_folder_relaxed = './pos_predicted_relaxed'
pos_save_folder_unrelaxed = './pos_predicted_unrelaxed'
os.makedirs(pos_save_folder_relaxed, exist_ok=True)
os.makedirs(pos_save_folder_unrelaxed, exist_ok=True)
mae_list = [] ; error_list = []; v_list = [] ; mae_list2 = [] ; vv_unrelaxed = [] ; vv_relaxed = []; vv_prediction = []
mae_list_unrelaxed = [] ; mae_list_relaxed = []; error_list_unrelaxed = [] ; error_list_relaxed = [] ; mae_list_previous = []
for i, (input,target_cell,target_pos,cif_ids) in enumerate(tqdm(test_loader)):
with torch.no_grad():
input_var = (Variable(input[0].cuda()),
Variable(input[1].cuda()),
input[2].cuda(),
input[3].cuda(),
input[4].cuda(),
input[5].cuda())
unrelaxed_feature = gcn.Encoding(*input_var)
unrelaxed_feature = Variable(unrelaxed_feature,volatile=True)
h_fea_len = unrelaxed_feature.shape[-1]
feature_length = int(h_fea_len**0.5)
feature_delta = generator(unrelaxed_feature.reshape(-1,1,feature_length,feature_length)).reshape(-1,h_fea_len)
translated_feature = unrelaxed_feature - feature_delta
atoms_fea = torch.cat((input[0],input[7]),dim=-1)
input_var2 = (Variable(atoms_fea.cuda()),
Variable(input[1].cuda()),
input[2].cuda(),
input[3].cuda(),
input[4].cuda(),
input[5].cuda(),
unrelaxed_feature,
translated_feature,
input[9][:,:9].cuda()
) ; target = target_pos
batch_size = target.shape[0]
target_var = Variable(target.cuda(), volatile=True)
output = model(*input_var2)
mae_error = mae(output.data.cpu(),target)
mae_list.append(mae_error)
# print(input[8][:,:9], target, output)
crystal_atom_idx = input[5].numpy()
#unrelaxed_pos = input[0][:,-3:]
unrelaxed_pos = input[7]
# print(unrelaxed_pos, unrelaxed_pos.shape)
count = 0
idxs, nums = np.unique(crystal_atom_idx,return_counts=True)
output = output.detach().cpu().numpy()
for i in idxs:
num_i = nums[i]
pos_unrelaxed = unrelaxed_pos[count:count+num_i].numpy()
pos_relaxed = target[count:count+num_i].numpy() + pos_unrelaxed
pos_prediction = output[count:count+num_i] + pos_unrelaxed
# pos_prediction = pos_unrelaxed #not changed
count += num_i
if '_unrelaxed' in cif_ids[i]:
name = cif_ids[i].split('_unrelaxed')[0]+'_pos.npy'
mae_list_unrelaxed.append(mae(torch.Tensor(pos_prediction),torch.Tensor(pos_relaxed)))
error_list_unrelaxed.append(np.mean(np.mean(abs((pos_relaxed - pos_unrelaxed) - (pos_prediction - pos_unrelaxed)),axis=-1),axis=-1))
mae_list_previous.append(mae(torch.Tensor(pos_unrelaxed),torch.Tensor(pos_relaxed)))
np.save(pos_save_folder_unrelaxed+'/'+name,pos_prediction)
else:
name = cif_ids[i].split('_relaxed')[0]+'_pos.npy'
mae_list_relaxed.append(mae(torch.Tensor(pos_prediction),torch.Tensor(pos_relaxed)))
error_list_relaxed.append(np.mean(np.mean(abs((pos_relaxed - pos_unrelaxed) - (pos_prediction - pos_unrelaxed)),axis=-1),axis=-1))
np.save(pos_save_folder_relaxed+'/'+name,pos_prediction)
print('Total MAE : ',np.mean(np.array(mae_list)))
print('Unrelaxed MAE : ',np.mean(np.array(mae_list_unrelaxed)))
print('Relaxed MAE : ',np.mean(np.array(mae_list_relaxed)))
print('Previous MAE : ',np.mean(np.array(mae_list_previous)))
error_list_unrelaxed = np.array(error_list_unrelaxed)
error_list_relaxed = np.array(error_list_relaxed)
print(error_list_unrelaxed.shape)
np.save('pos_error_list_unrelaxed.npy',error_list_unrelaxed)
np.save('pos_error_list_relaxed.npy',error_list_relaxed)
def mae(prediction, target):
return torch.mean(torch.abs(target - prediction))
if __name__ == '__main__':
main()