Commit dda202ee authored by Marco Cristoforetti's avatar Marco Cristoforetti
Browse files

net

parent 9a624e84
......@@ -1042,7 +1042,19 @@
"Epoch 677 time = 10.84, tr_rmse = 10.82458, val_rmse = 10.20095, ts_rmse = 10.15267, tr_c = 0.34388, val_c = 0.25613, ts_c = 0.25155\n",
"Epoch 678 time = 9.91, tr_rmse = 10.52713, val_rmse = 9.35514, ts_rmse = 9.43322, tr_c = 0.34206, val_c = 0.25281, ts_c = 0.25070\n",
"Epoch 679 time = 9.93, tr_rmse = 9.97220, val_rmse = 9.34884, ts_rmse = 9.42746, tr_c = 0.33628, val_c = 0.25170, ts_c = 0.24863\n",
"Epoch 680 time = 9.72, tr_rmse = 10.60701, val_rmse = 9.84598, ts_rmse = 9.84561, tr_c = 0.34229, val_c = 0.25474, ts_c = 0.25055\n"
"Epoch 680 time = 9.72, tr_rmse = 10.60701, val_rmse = 9.84598, ts_rmse = 9.84561, tr_c = 0.34229, val_c = 0.25474, ts_c = 0.25055\n",
"Epoch 681 time = 9.70, tr_rmse = 10.31652, val_rmse = 9.54120, ts_rmse = 9.50236, tr_c = 0.34368, val_c = 0.25264, ts_c = 0.24780\n",
"Epoch 682 time = 9.60, tr_rmse = 10.18568, val_rmse = 9.48085, ts_rmse = 9.50323, tr_c = 0.34025, val_c = 0.25029, ts_c = 0.24641\n",
"Epoch 683 time = 9.63, tr_rmse = 10.51845, val_rmse = 10.05414, ts_rmse = 10.00518, tr_c = 0.36263, val_c = 0.27014, ts_c = 0.26403\n",
"Epoch 684 time = 9.45, tr_rmse = 10.85813, val_rmse = 10.05507, ts_rmse = 9.93873, tr_c = 0.36799, val_c = 0.27127, ts_c = 0.26504\n",
"Epoch 685 time = 9.48, tr_rmse = 10.33319, val_rmse = 9.31879, ts_rmse = 9.42276, tr_c = 0.34676, val_c = 0.25698, ts_c = 0.25518\n",
"Epoch 686 time = 9.25, tr_rmse = 10.67490, val_rmse = 10.28641, ts_rmse = 10.36100, tr_c = 0.34734, val_c = 0.25926, ts_c = 0.25605\n",
"Epoch 687 time = 9.12, tr_rmse = 10.01224, val_rmse = 9.47536, ts_rmse = 9.52291, tr_c = 0.33570, val_c = 0.25012, ts_c = 0.24672\n",
"Epoch 688 time = 9.28, tr_rmse = 10.85141, val_rmse = 10.14378, ts_rmse = 10.03594, tr_c = 0.37271, val_c = 0.27083, ts_c = 0.26359\n",
"Epoch 689 time = 9.05, tr_rmse = 10.32716, val_rmse = 9.18625, ts_rmse = 9.25313, tr_c = 0.34393, val_c = 0.25228, ts_c = 0.25067\n",
"Epoch 690 time = 9.24, tr_rmse = 10.02979, val_rmse = 9.18719, ts_rmse = 9.24496, tr_c = 0.33852, val_c = 0.25201, ts_c = 0.24832\n",
"Epoch 691 time = 9.44, tr_rmse = 10.37609, val_rmse = 9.68697, ts_rmse = 9.77203, tr_c = 0.33653, val_c = 0.24952, ts_c = 0.24673\n",
"Epoch 692 time = 9.12, tr_rmse = 10.30415, val_rmse = 9.68349, ts_rmse = 9.64272, tr_c = 0.34494, val_c = 0.25910, ts_c = 0.25442\n"
]
}
],
......
%% Cell type:code id: tags:
``` python
%matplotlib inline
import matplotlib.pyplot as plt
import sys; sys.path.append('../DST')
import os
from DST.config import data_path
import pandas as pd
import numpy as np
import seaborn as sns; sns.set(style="whitegrid", font_scale=1.3)
import torch
import torch.nn as nn
import time
import math
import torch.utils.data as utils_data
import torch.nn.functional as F
import datetime
```
%% Cell type:code id: tags:
``` python
torch.manual_seed(21894)
np.random.seed(21894)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
```
%% Cell type:code id: tags:
``` python
BEFORE = 12
AFTER = 12
```
%% Cell type:code id: tags:
``` python
dst_data = pd.read_pickle(os.path.join(data_path,'dst.pkl'))
dst_data['ora_round'] = dst_data.ora.apply(lambda x:int(x.split(':')[0]))
dati_agg = dst_data.groupby(['data','ora_round']).agg({
'BX': np.mean,
'BY': np.mean,
'BZ': np.mean,
'FLOW_SPEED': np.mean,
'PROTON_DENSITY': np.mean,
'TEMPERATURE': np.mean,
'PRESSION': np.mean,
'ELETTRIC': np.mean,
'y': np.mean})
dati_agg.reset_index(inplace=True)
dati_agg.sort_values(by = ['data','ora_round'],inplace=True)
dataset = dati_agg.drop(columns = ['data','ora_round']).values
dataset = torch.from_numpy(np.hstack([np.arange(len(dataset)).reshape([-1,1]),dataset]))
last_date_train = dati_agg[dati_agg.data <= datetime.datetime(2008,12,31)].index[-1]
len_valid_test = (len(dataset) - last_date_train)/2
last_date_train/len(dataset), len_valid_test/len(dataset)
data_in = dataset.unfold(0, BEFORE, 1).transpose(2,1)
data_out = dataset[BEFORE:].unfold(0, AFTER, 1).transpose(2,1)
data_in = data_in[:data_out.size(0)]
data_out = data_out[:,:,-1]
data_in.size(), data_out.size()
```
%%%% Output: execute_result
(torch.Size([261794, 12, 10]), torch.Size([261794, 12]))
%% Cell type:code id: tags:
``` python
where_not_nan_in = ~torch.isnan(data_in).any(2, keepdim=True).any(1, keepdim=True).reshape(-1)
data_in = data_in[where_not_nan_in]
data_out = data_out[where_not_nan_in]
where_not_nan_out = ~torch.isnan(data_out).any(1, keepdim=True).reshape(-1)
data_in = data_in[where_not_nan_out]
data_out = data_out[where_not_nan_out]
last_train = np.where(data_in[:,0,0] <= last_date_train)[0][-1] + 1
data_in = data_in[:, :, 1:]
#len_tr = int(data_in.size(0) * 0.6)
n_channels = data_in.size(2)
```
%% Cell type:code id: tags:
``` python
class MinMaxScaler():
"""
Transform features by scaling each feature to a given range
Features in the last dim
The transformation is given by::
X_std = (X - X.min(axis=0)) / (X.max(axis=0) - X.min(axis=0))
X_scaled = X_std * (max - min) + min
where min, max = feature_range.
"""
def __init__(self, feature_range=(0,1)):
self.feature_range = feature_range
def fit(self, X):
X_size = X.size()
X = X.reshape(-1, X_size[-1])
data_min = X.min(axis=0).values
data_max = X.max(axis=0).values
data_range = data_max - data_min
self.scale_ = ((self.feature_range[1] - self.feature_range[0]) / data_range)
self.min_ = self.feature_range[0] - data_min * self.scale_
self.data_min_ = data_min
self.data_max_ = data_max
self.data_range_ = data_range
X = X.reshape(X_size)
return self
def transform(self, X):
X *= self.scale_
X += self.min_
return X
def inverse_transform(self, X):
X -= self.min_
X /= self.scale_
return X
```
%% Cell type:code id: tags:
``` python
mmScaler = MinMaxScaler((0.1, .9))
mmScaler.fit(data_in[:last_train])
data_in_scaled = data_in.clone()
data_in_scaled = mmScaler.transform(data_in_scaled)
mm_scaler_out = MinMaxScaler((0.1, .9))
mm_scaler_out.fit(data_in[:last_train, :, -1].reshape(-1, data_in.size(1), 1))
data_out_scaled = data_out.clone()
data_out_scaled = mm_scaler_out.transform(data_out_scaled)
```
%% Cell type:code id: tags:
``` python
dst_levels = [-20,-50,-100]
data_out_c = data_out.clone()
data_out_c[np.where(data_out_c >= dst_levels[0])] = 0
data_out_c[np.where((data_out_c < dst_levels[0]) & (data_out_c >= dst_levels[1]))] = 1
data_out_c[np.where((data_out_c < dst_levels[1]) & (data_out_c >= dst_levels[2]))] = 2
data_out_c[np.where((data_out_c < dst_levels[2]))] = 3
```
%% Cell type:code id: tags:
``` python
class Dataset(utils_data.Dataset):
def __init__(self, dataset_in, dataset_out, dataset_out_c, weights):
self.dataset_in = dataset_in
self.dataset_out = dataset_out
self.dataset_out_c = dataset_out_c
self.weights = weights
def __len__(self):
return self.dataset_in.size(0)
def __getitem__(self, idx):
din_src = self.dataset_in[idx]
dout = self.dataset_out[idx]
dout_c = self.dataset_out_c[idx]
ww = self.weights[idx]
return din_src, dout, dout_c, ww
```
%% Cell type:code id: tags:
``` python
ixs_valid_test = np.arange(int(len_valid_test)) + last_train
np.random.shuffle(ixs_valid_test)
ixs_valid = ixs_valid_test[::2]
ixs_test = ixs_valid_test[1::2]
```
%% Cell type:code id: tags:
``` python
dst_min = data_out[:last_train].min(axis=1).values.flatten()
bins = [dst_min.min() - 10] + list(np.arange(-300, dst_min.max() + 10, 10))
h, b = np.histogram(dst_min, bins=bins)
if len(np.argwhere(h == 0)) > 0:
bins = np.delete(bins, np.argwhere(h == 0)[0] + 1)
h, b = np.histogram(dst_min, bins=bins)
w = h.max()/h
def fix_weight(dst_v):
pos = np.argwhere(np.abs(b - dst_v) == np.abs((b - dst_v)).min())[0,0]
if dst_v - b[pos] < 0:
pos = pos-1
# return w[pos]/h.max()
return np.sqrt(w[pos]/h.max())
fix_weight_v = np.vectorize(fix_weight)
weights = fix_weight_v(dst_min)
sampler = torch.utils.data.sampler.WeightedRandomSampler(weights, num_samples= len(dst_min))
BATCH_SIZE=256
dataset_tr = Dataset(data_in_scaled[:last_train], data_out_scaled[:last_train], data_out_c[:last_train], weights)
data_loader_tr = utils_data.DataLoader(dataset_tr, batch_size=BATCH_SIZE, num_workers = 4, shuffle=False, sampler = sampler)
# data_loader_tr = utils_data.DataLoader(dataset_tr, batch_size=BATCH_SIZE, num_workers = 4, shuffle=True)
```
%% Cell type:code id: tags:
``` python
class DSTnet(nn.Module):
def __init__(self, nvars, nhidden_i, nhidden_o, n_out_i, before, after):
super().__init__()
self.nvars = nvars
self.nhidden_i = nhidden_i
self.nhidden_o = nhidden_o
self.before = before
self.after = after
self.n_out_i = n_out_i
self.lstm = nn.LSTM(self.nvars, self.n_out_i, self.nhidden_i, batch_first=True)
self.first_merged_layer = self.n_out_i * self.before
self.bn1 = nn.BatchNorm1d(num_features=self.first_merged_layer)
self.linear_o_1 = nn.Linear(self.first_merged_layer, self.nhidden_o)
self.ln1 = nn.LayerNorm(self.nhidden_o )
self.linear_o_2 = nn.Linear(self.nhidden_o, self.nhidden_o)
self.linear_o_3 = nn.Linear(self.nhidden_o, self.nhidden_o // 2)
self.linear_o_4_c = nn.Linear(self.nhidden_o // 2, self.after * 4)
self.linear_o_4_r = nn.Linear(4, 16)
self.linear_o_4b_r = nn.Linear(4, 4)
self.linear_o_5_r = nn.Linear(16, 1)
def init_hidden(self, batch_size):
hidden = torch.randn(self.nhidden_i, batch_size, self.n_out_i).to(device)
cell = torch.randn(self.nhidden_i, batch_size, self.n_out_i).to(device)
return (hidden, cell)
def forward(self, x0):
self.hidden = self.init_hidden(x0.size(0))
x = self.lstm(x0, self.hidden)[0].reshape(x0.shape[0], -1)
x = self.bn1(x)
# x = F.relu(x)
x = F.relu(self.linear_o_1(x))
# x = self.ln1(x)
x = F.dropout(x, 0.2, training=self.training)
x = F.relu(self.linear_o_2(x))
x = F.dropout(x, 0.2, training=self.training)
x = F.relu(self.linear_o_3(x))
x = F.dropout(x, 0.2, training=self.training)
x1 = self.linear_o_4_c(x)
x2 = F.relu(self.linear_o_4_r(x1.view(-1, 4)))
# x2 = F.relu(self.linear_o_4b_r(x2))
x2 = self.linear_o_5_r(x2)
x2 = x2.reshape(x0.size(0), self.after)
x1 = x1.reshape(x0.size(0) * self.after, 4)
return x2, x1
```
%% Cell type:code id: tags:
``` python
aa = data_out_c[:last_train]
weights_c = torch.tensor([len(aa[aa==0])/len(aa[aa==0]), len(aa[aa==0])/len(aa[aa==1]), len(aa[aa==0])/len(aa[aa==2]), len(aa[aa==0])/len(aa[aa==3])]).to(device).sqrt()
```
%% Cell type:code id: tags:
``` python
loss_f = nn.L1Loss()
loss_mse = nn.MSELoss(reduction='none')
loss_fc= nn.CrossEntropyLoss()
# loss_fc= nn.CrossEntropyLoss(weight = weights_c)
nhidden_i = 2
nhidden_o = 96
n_out_i = 8
before = BEFORE
nvars = data_in_scaled.shape[-1]
dst_net = DSTnet(nvars, nhidden_i, nhidden_o, n_out_i, before, AFTER).to(device)
print(dst_net)
num_epochs = 2000
lr = 1e-4
optimizer = torch.optim.Adam(dst_net.parameters(), lr=lr)#, weight_decay=1e-5)
history_tr = np.zeros((num_epochs, 3))
history_valid = np.zeros((num_epochs, 3))
history_ts = np.zeros((num_epochs, 3))
for epoch in range(num_epochs):
start_time = time.time()
for i, batch in enumerate(data_loader_tr):
x = batch[0].float().to(device)
y_r = batch[1].float().to(device)
y_c = batch[2].flatten().long().to(device)
w = batch[3].to(device)
optimizer.zero_grad()
dst_net.train()
out_r, out_c = dst_net(x)
loss_r = loss_f(out_r, y_r)
loss_c = loss_fc(out_c, y_c)
# loss = (loss_r * w).mean() + 10 * loss_c
loss = (loss_r).mean() + loss_c
loss.backward()
optimizer.step()
dst_net.eval()
out_r, out_c = dst_net(data_in_scaled[:last_train].to(device).float())
loss_tr = np.sqrt(loss_mse(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[:last_train].to(device).float()).mean().item())
loss_mae_tr = loss_f(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[:last_train].to(device).float()).item()
loss_c_tr = loss_fc(out_c, data_out_c[:last_train].flatten().long().to(device)).item()
out_r, out_c = dst_net(data_in_scaled[ixs_valid].to(device).float())
loss_val = np.sqrt(loss_mse(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[ixs_valid].to(device).float()).mean().item())
loss_mae_val = loss_f(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[ixs_valid].to(device).float()).item()
loss_c_val = loss_fc(out_c, data_out_c[ixs_valid].flatten().long().to(device)).item()
out_r, out_c = dst_net(data_in_scaled[ixs_test].to(device).float())
loss_ts = np.sqrt(loss_mse(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[ixs_test].to(device).float()).mean().item())
loss_mae_ts = loss_f(mm_scaler_out.inverse_transform(out_r.cpu().clone()).to(device), data_out[ixs_test].to(device).float()).item()
loss_c_ts = loss_fc(out_c, data_out_c[ixs_test].flatten().long().to(device)).item()
history_tr[epoch] = [loss_tr, loss_mae_tr, loss_c_tr]
history_valid[epoch] = [loss_val, loss_mae_val, loss_c_val]
history_ts[epoch] = [loss_ts, loss_mae_ts, loss_c_ts]
epoch_time = time.time() - start_time
if (epoch % 1 == 0):
print('Epoch %d time = %.2f, tr_rmse = %0.5f, val_rmse = %0.5f, ts_rmse = %0.5f, tr_c = %.5f, val_c = %.5f, ts_c = %.5f' %
(epoch, epoch_time, loss_tr, loss_val, loss_ts, loss_c_tr, loss_c_val, loss_c_ts))
```
%%%% Output: stream
DSTnet(
(lstm): LSTM(9, 8, num_layers=2, batch_first=True)
(bn1): BatchNorm1d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(linear_o_1): Linear(in_features=96, out_features=96, bias=True)
(ln1): LayerNorm((96,), eps=1e-05, elementwise_affine=True)
(linear_o_2): Linear(in_features=96, out_features=96, bias=True)
(linear_o_3): Linear(in_features=96, out_features=48, bias=True)
(linear_o_4_c): Linear(in_features=48, out_features=48, bias=True)
(linear_o_4_r): Linear(in_features=4, out_features=16, bias=True)
(linear_o_4b_r): Linear(in_features=4, out_features=4, bias=True)
(linear_o_5_r): Linear(in_features=16, out_features=1, bias=True)
)
Epoch 0 time = 8.30, tr_rmse = 66.15217, val_rmse = 67.77470, ts_rmse = 67.68658, tr_c = 0.60882, val_c = 0.43454, ts_c = 0.43171
Epoch 1 time = 8.55, tr_rmse = 45.34703, val_rmse = 34.77569, ts_rmse = 34.57838, tr_c = 0.49833, val_c = 0.34987, ts_c = 0.34553
Epoch 2 time = 9.85, tr_rmse = 32.92374, val_rmse = 23.71364, ts_rmse = 23.49459, tr_c = 0.47043, val_c = 0.33199, ts_c = 0.32764
Epoch 3 time = 9.51, tr_rmse = 20.39029, val_rmse = 14.39769, ts_rmse = 14.35967, tr_c = 0.43625, val_c = 0.30474, ts_c = 0.30048
Epoch 4 time = 9.28, tr_rmse = 17.22628, val_rmse = 12.43394, ts_rmse = 12.47325, tr_c = 0.42075, val_c = 0.29253, ts_c = 0.28882
Epoch 5 time = 9.26, tr_rmse = 16.29019, val_rmse = 12.05787, ts_rmse = 12.07865, tr_c = 0.41807, val_c = 0.29102, ts_c = 0.28699
Epoch 6 time = 10.10, tr_rmse = 15.42468, val_rmse = 11.51328, ts_rmse = 11.60580, tr_c = 0.41134, val_c = 0.28505, ts_c = 0.28161
Epoch 7 time = 9.70, tr_rmse = 15.92914, val_rmse = 11.61244, ts_rmse = 11.63528, tr_c = 0.42003, val_c = 0.29098, ts_c = 0.28642
Epoch 8 time = 9.86, tr_rmse = 15.27134, val_rmse = 11.56234, ts_rmse = 11.60855, tr_c = 0.40577, val_c = 0.28070, ts_c = 0.27646
Epoch 9 time = 9.47, tr_rmse = 15.06807, val_rmse = 11.24741, ts_rmse = 11.19969, tr_c = 0.42280, val_c = 0.29420, ts_c = 0.28966
Epoch 10 time = 9.28, tr_rmse = 14.28239, val_rmse = 10.65062, ts_rmse = 10.68373, tr_c = 0.40186, val_c = 0.27974, ts_c = 0.27544
Epoch 11 time = 9.41, tr_rmse = 14.16488, val_rmse = 10.77011, ts_rmse = 10.85400, tr_c = 0.40196, val_c = 0.27817, ts_c = 0.27376
Epoch 12 time = 9.20, tr_rmse = 14.20300, val_rmse = 10.70767, ts_rmse = 10.67010, tr_c = 0.40315, val_c = 0.27866, ts_c = 0.27401
Epoch 13 time = 9.10, tr_rmse = 13.86771, val_rmse = 10.71816, ts_rmse = 10.74507, tr_c = 0.39955, val_c = 0.27742, ts_c = 0.27260
Epoch 14 time = 9.28, tr_rmse = 14.28546, val_rmse = 10.77270, ts_rmse = 10.72609, tr_c = 0.42552, val_c = 0.29723, ts_c = 0.29143
Epoch 15 time = 9.15, tr_rmse = 13.66945, val_rmse = 10.36490, ts_rmse = 10.37500, tr_c = 0.40354, val_c = 0.27991, ts_c = 0.27519
Epoch 16 time = 9.09, tr_rmse = 13.35821, val_rmse = 10.24788, ts_rmse = 10.26383, tr_c = 0.41342, val_c = 0.28998, ts_c = 0.28469
Epoch 17 time = 9.07, tr_rmse = 13.22511, val_rmse = 10.03119, ts_rmse = 10.05284, tr_c = 0.40312, val_c = 0.28198, ts_c = 0.27677
Epoch 18 time = 8.93, tr_rmse = 13.00985, val_rmse = 10.09758, ts_rmse = 10.11566, tr_c = 0.39448, val_c = 0.27303, ts_c = 0.26756
Epoch 19 time = 9.03, tr_rmse = 13.33757, val_rmse = 10.92243, ts_rmse = 10.96307, tr_c = 0.39517, val_c = 0.27227, ts_c = 0.26754
Epoch 20 time = 8.86, tr_rmse = 12.85982, val_rmse = 10.11336, ts_rmse = 10.15719, tr_c = 0.39503, val_c = 0.27272, ts_c = 0.26776
Epoch 21 time = 9.33, tr_rmse = 12.68230, val_rmse = 9.66850, ts_rmse = 9.68617, tr_c = 0.39521, val_c = 0.27419, ts_c = 0.26926
Epoch 22 time = 9.19, tr_rmse = 12.62882, val_rmse = 9.74624, ts_rmse = 9.75907, tr_c = 0.39376, val_c = 0.27283, ts_c = 0.26780
Epoch 23 time = 9.02, tr_rmse = 12.79253, val_rmse = 9.73440, ts_rmse = 9.74277, tr_c = 0.41072, val_c = 0.28569, ts_c = 0.27952
Epoch 24 time = 9.03, tr_rmse = 12.58978, val_rmse = 9.60329, ts_rmse = 9.62610, tr_c = 0.40625, val_c = 0.28374, ts_c = 0.27803
Epoch 25 time = 8.89, tr_rmse = 12.47839, val_rmse = 9.89261, ts_rmse = 9.95068, tr_c = 0.39415, val_c = 0.27117, ts_c = 0.26636
Epoch 26 time = 8.69, tr_rmse = 12.35089, val_rmse = 9.54690, ts_rmse = 9.58202, tr_c = 0.39700, val_c = 0.27307, ts_c = 0.26840
Epoch 27 time = 8.89, tr_rmse = 12.87943, val_rmse = 10.78383, ts_rmse = 10.86031, tr_c = 0.39292, val_c = 0.27149, ts_c = 0.26687
Epoch 28 time = 8.79, tr_rmse = 12.32241, val_rmse = 9.42373, ts_rmse = 9.45090, tr_c = 0.39891, val_c = 0.27675, ts_c = 0.27116
Epoch 29 time = 9.07, tr_rmse = 12.27768, val_rmse = 9.64752, ts_rmse = 9.67976, tr_c = 0.39592, val_c = 0.27355, ts_c = 0.26881
Epoch 30 time = 8.90, tr_rmse = 12.22796, val_rmse = 9.48093, ts_rmse = 9.50684, tr_c = 0.39538, val_c = 0.27190, ts_c = 0.26628
Epoch 31 time = 8.97, tr_rmse = 12.07956, val_rmse = 9.37767, ts_rmse = 9.38912, tr_c = 0.38751, val_c = 0.26672, ts_c = 0.26150
Epoch 32 time = 9.01, tr_rmse = 12.18557, val_rmse = 9.39537, ts_rmse = 9.39264, tr_c = 0.39340, val_c = 0.27105, ts_c = 0.26546
Epoch 33 time = 8.82, tr_rmse = 12.10100, val_rmse = 9.62596, ts_rmse = 9.66562, tr_c = 0.38421, val_c = 0.26576, ts_c = 0.26051
Epoch 34 time = 8.97, tr_rmse = 11.99623, val_rmse = 9.28614, ts_rmse = 9.29758, tr_c = 0.38842, val_c = 0.27000, ts_c = 0.26472
Epoch 35 time = 9.28, tr_rmse = 12.60344, val_rmse = 9.79549, ts_rmse = 9.77828, tr_c = 0.40781, val_c = 0.28407, ts_c = 0.27817
Epoch 36 time = 9.15, tr_rmse = 12.08264, val_rmse = 9.35097, ts_rmse = 9.34065, tr_c = 0.39115, val_c = 0.26964, ts_c = 0.26412
Epoch 37 time = 9.09, tr_rmse = 12.27970, val_rmse = 9.97233, ts_rmse = 10.01524, tr_c = 0.39641, val_c = 0.27374, ts_c = 0.26864
Epoch 38 time = 8.99, tr_rmse = 11.86919, val_rmse = 9.24856, ts_rmse = 9.26881, tr_c = 0.38292, val_c = 0.26391, ts_c = 0.25878
Epoch 39 time = 9.10, tr_rmse = 12.00048, val_rmse = 9.49151, ts_rmse = 9.50516, tr_c = 0.38864, val_c = 0.26742, ts_c = 0.26214
Epoch 40 time = 8.78, tr_rmse = 12.02184, val_rmse = 9.32621, ts_rmse = 9.29207, tr_c = 0.38662, val_c = 0.26728, ts_c = 0.26159
Epoch 41 time = 8.86, tr_rmse = 11.94749, val_rmse = 9.28574, ts_rmse = 9.28127, tr_c = 0.39902, val_c = 0.27585, ts_c = 0.27031
Epoch 42 time = 9.02, tr_rmse = 11.90864, val_rmse = 9.23759, ts_rmse = 9.21802, tr_c = 0.39061, val_c = 0.27168, ts_c = 0.26605
Epoch 43 time = 8.73, tr_rmse = 11.90221, val_rmse = 9.47429, ts_rmse = 9.48570, tr_c = 0.38642, val_c = 0.26661, ts_c = 0.26102
Epoch 44 time = 8.91, tr_rmse = 11.80015, val_rmse = 9.36381, ts_rmse = 9.36680, tr_c = 0.38052, val_c = 0.26179, ts_c = 0.25669
Epoch 45 time = 8.89, tr_rmse = 11.92069, val_rmse = 9.25713, ts_rmse = 9.24335, tr_c = 0.39986, val_c = 0.27812, ts_c = 0.27268
Epoch 46 time = 8.90, tr_rmse = 12.36547, val_rmse = 9.57037, ts_rmse = 9.55312, tr_c = 0.42695, val_c = 0.29653, ts_c = 0.29037
Epoch 47 time = 9.04, tr_rmse = 12.03019, val_rmse = 9.41958, ts_rmse = 9.39346, tr_c = 0.37262, val_c = 0.25566, ts_c = 0.25038
Epoch 48 time = 9.07, tr_rmse = 12.03370, val_rmse = 9.81657, ts_rmse = 9.83969, tr_c = 0.38703, val_c = 0.26624, ts_c = 0.26105
Epoch 49 time = 8.90, tr_rmse = 11.70577, val_rmse = 9.12597, ts_rmse = 9.11509, tr_c = 0.37881, val_c = 0.26208, ts_c = 0.25685
Epoch 50 time = 8.93, tr_rmse = 11.77391, val_rmse = 9.18155, ts_rmse = 9.14606, tr_c = 0.38487, val_c = 0.26677, ts_c = 0.26136
Epoch 51 time = 8.85, tr_rmse = 11.69276, val_rmse = 9.21197, ts_rmse = 9.19295, tr_c = 0.38751, val_c = 0.26781, ts_c = 0.26140
Epoch 52 time = 8.92, tr_rmse = 11.71989, val_rmse = 9.14561, ts_rmse = 9.15013, tr_c = 0.38444, val_c = 0.26665, ts_c = 0.26100
Epoch 53 time = 8.70, tr_rmse = 11.74966, val_rmse = 9.48791, ts_rmse = 9.48643, tr_c = 0.37861, val_c = 0.26177, ts_c = 0.25607
Epoch 54 time = 9.17, tr_rmse = 12.22945, val_rmse = 9.53070, ts_rmse = 9.53073, tr_c = 0.41557, val_c = 0.29334, ts_c = 0.28742
Epoch 55 time = 9.10, tr_rmse = 11.81570, val_rmse = 9.28387, ts_rmse = 9.23942, tr_c = 0.38242, val_c = 0.26475, ts_c = 0.25844
Epoch 56 time = 9.44, tr_rmse = 12.43880, val_rmse = 9.66847, ts_rmse = 9.67556, tr_c = 0.42866, val_c = 0.30117, ts_c = 0.29511
Epoch 57 time = 9.08, tr_rmse = 11.74656, val_rmse = 9.56894, ts_rmse = 9.58574, tr_c = 0.37004, val_c = 0.25495, ts_c = 0.24954
Epoch 58 time = 9.13, tr_rmse = 12.13151, val_rmse = 9.59704, ts_rmse = 9.51374, tr_c = 0.44730, val_c = 0.32045, ts_c = 0.31308
Epoch 59 time = 9.15, tr_rmse = 11.90883, val_rmse = 9.61407, ts_rmse = 9.62709, tr_c = 0.41531, val_c = 0.28979, ts_c = 0.28361
Epoch 60 time = 8.99, tr_rmse = 11.72765, val_rmse = 9.39450, ts_rmse = 9.37840, tr_c = 0.38919, val_c = 0.26902, ts_c = 0.26306
Epoch 61 time = 9.42, tr_rmse = 11.79716, val_rmse = 9.24748, ts_rmse = 9.24074, tr_c = 0.37563, val_c = 0.25920, ts_c = 0.25385
Epoch 62 time = 9.19, tr_rmse = 11.65213, val_rmse = 9.14660, ts_rmse = 9.14051, tr_c = 0.36701, val_c = 0.25233, ts_c = 0.24722
Epoch 63 time = 8.82, tr_rmse = 11.57939, val_rmse = 9.32544, ts_rmse = 9.32719, tr_c = 0.36803, val_c = 0.25371, ts_c = 0.24825
Epoch 64 time = 9.08, tr_rmse = 11.60288, val_rmse = 9.14625, ts_rmse = 9.17145, tr_c = 0.37526, val_c = 0.25798, ts_c = 0.25306
Epoch 65 time = 9.02, tr_rmse = 11.59354, val_rmse = 9.37283, ts_rmse = 9.36926, tr_c = 0.37211, val_c = 0.25722, ts_c = 0.25151
Epoch 66 time = 9.05, tr_rmse = 11.60345, val_rmse = 9.15976, ts_rmse = 9.16513, tr_c = 0.37865, val_c = 0.26200, ts_c = 0.25623
Epoch 67 time = 8.91, tr_rmse = 11.62165, val_rmse = 9.39010, ts_rmse = 9.35768, tr_c = 0.38792, val_c = 0.27148, ts_c = 0.26506
Epoch 68 time = 8.96, tr_rmse = 11.54365, val_rmse = 9.07304, ts_rmse = 9.06762, tr_c = 0.36832, val_c = 0.25429, ts_c = 0.24903
Epoch 69 time = 9.05, tr_rmse = 12.08266, val_rmse = 9.53859, ts_rmse = 9.47496, tr_c = 0.38516, val_c = 0.26756, ts_c = 0.26184
Epoch 70 time = 8.98, tr_rmse = 11.60947, val_rmse = 9.41017, ts_rmse = 9.43446, tr_c = 0.36459, val_c = 0.25103, ts_c = 0.24577
Epoch 71 time = 8.99, tr_rmse = 12.38703, val_rmse = 9.91685, ts_rmse = 9.89209, tr_c = 0.44415, val_c = 0.31538, ts_c = 0.30876
Epoch 72 time = 9.07, tr_rmse = 11.84461, val_rmse = 9.31832, ts_rmse = 9.29477, tr_c = 0.38731, val_c = 0.27047, ts_c = 0.26424
Epoch 73 time = 8.98, tr_rmse = 11.65994, val_rmse = 9.17506, ts_rmse = 9.14670, tr_c = 0.37376, val_c = 0.25846, ts_c = 0.25253
Epoch 74 time = 8.98, tr_rmse = 11.91778, val_rmse = 9.92754, ts_rmse = 9.94171, tr_c = 0.37436, val_c = 0.25933, ts_c = 0.25397
Epoch 75 time = 9.05, tr_rmse = 11.58297, val_rmse = 9.47976, ts_rmse = 9.48462, tr_c = 0.37456, val_c = 0.26139, ts_c = 0.25552
Epoch 76 time = 8.97, tr_rmse = 12.20070, val_rmse = 10.42143, ts_rmse = 10.43249, tr_c = 0.37978, val_c = 0.26223, ts_c = 0.25672
Epoch 77 time = 8.94, tr_rmse = 11.97290, val_rmse = 9.61334, ts_rmse = 9.56335, tr_c = 0.37520, val_c = 0.26153, ts_c = 0.25545
Epoch 78 time = 8.91, tr_rmse = 11.50731, val_rmse = 9.11409, ts_rmse = 9.08390, tr_c = 0.37559, val_c = 0.26024, ts_c = 0.25448
Epoch 79 time = 8.99, tr_rmse = 11.92793, val_rmse = 9.39444, ts_rmse = 9.32600, tr_c = 0.39528, val_c = 0.27542, ts_c = 0.26897
Epoch 80 time = 8.85, tr_rmse = 11.66363, val_rmse = 9.28862, ts_rmse = 9.26887, tr_c = 0.39435, val_c = 0.27387, ts_c = 0.26765
Epoch 81 time = 8.88, tr_rmse = 11.66350, val_rmse = 9.42915, ts_rmse = 9.38685, tr_c = 0.38762, val_c = 0.26971, ts_c = 0.26330
Epoch 82 time = 9.02, tr_rmse = 11.77905, val_rmse = 9.39174, ts_rmse = 9.37781, tr_c = 0.36908, val_c = 0.25455, ts_c = 0.24931
Epoch 83 time = 9.00, tr_rmse = 11.41212, val_rmse = 9.12332, ts_rmse = 9.11387, tr_c = 0.36584, val_c = 0.25280, ts_c = 0.24736
Epoch 84 time = 9.04, tr_rmse = 12.75888, val_rmse = 9.93278, ts_rmse = 9.88806, tr_c = 0.45938, val_c = 0.32963, ts_c = 0.32325
Epoch 85 time = 8.99, tr_rmse = 11.40654, val_rmse = 9.12228, ts_rmse = 9.11890, tr_c = 0.36839, val_c = 0.25631, ts_c = 0.25097
Epoch 86 time = 9.00, tr_rmse = 11.47390, val_rmse = 9.22761, ts_rmse = 9.25403, tr_c = 0.36359, val_c = 0.25184, ts_c = 0.24665
Epoch 87 time = 8.91, tr_rmse = 11.72040, val_rmse = 9.54289, ts_rmse = 9.52387, tr_c = 0.38241, val_c = 0.26678, ts_c = 0.26083
Epoch 88 time = 9.12, tr_rmse = 11.41654, val_rmse = 9.36794, ts_rmse = 9.36425, tr_c = 0.36737, val_c = 0.25590, ts_c = 0.25016
Epoch 89 time = 9.34, tr_rmse = 11.70840, val_rmse = 9.29624, ts_rmse = 9.25334, tr_c = 0.38543, val_c = 0.26777, ts_c = 0.26181
Epoch 90 time = 9.04, tr_rmse = 11.59987, val_rmse = 9.56446, ts_rmse = 9.58170, tr_c = 0.37019, val_c = 0.25706, ts_c = 0.25183
Epoch 91 time = 9.05, tr_rmse = 11.40675, val_rmse = 9.27711, ts_rmse = 9.31550, tr_c = 0.36160, val_c = 0.25018, ts_c = 0.24490
Epoch 92 time = 8.94, tr_rmse = 11.75237, val_rmse = 9.49403, ts_rmse = 9.43998, tr_c = 0.41684, val_c = 0.29383, ts_c = 0.28650
Epoch 93 time = 9.17, tr_rmse = 11.37276, val_rmse = 9.14449, ts_rmse = 9.15320, tr_c = 0.36473, val_c = 0.25334, ts_c = 0.24793
Epoch 94 time = 8.87, tr_rmse = 11.95308, val_rmse = 9.45088, ts_rmse = 9.42307, tr_c = 0.40041, val_c = 0.28421, ts_c = 0.27819
Epoch 95 time = 8.89, tr_rmse = 11.37392, val_rmse = 9.14861, ts_rmse = 9.13176, tr_c = 0.36883, val_c = 0.25685, ts_c = 0.25115
Epoch 96 time = 8.90, tr_rmse = 11.49933, val_rmse = 9.20025, ts_rmse = 9.17753, tr_c = 0.37185, val_c = 0.25831, ts_c = 0.25284
Epoch 97 time = 9.05, tr_rmse = 11.37629, val_rmse = 9.36312, ts_rmse = 9.35513, tr_c = 0.36169, val_c = 0.25128, ts_c = 0.24576
Epoch 98 time = 8.92, tr_rmse = 11.98322, val_rmse = 9.52955, ts_rmse = 9.49798, tr_c = 0.41017, val_c = 0.28982, ts_c = 0.28365
Epoch 99 time = 9.03, tr_rmse = 11.98974, val_rmse = 9.60245, ts_rmse = 9.55691, tr_c = 0.41787, val_c = 0.29465, ts_c = 0.28807
Epoch 100 time = 8.99, tr_rmse = 11.61499, val_rmse = 9.75127, ts_rmse = 9.73488, tr_c = 0.36950, val_c = 0.25695, ts_c = 0.25128
Epoch 101 time = 9.25, tr_rmse = 11.80067, val_rmse = 9.38467, ts_rmse = 9.37057, tr_c = 0.39493, val_c = 0.27800, ts_c = 0.27194
Epoch 102 time = 9.21, tr_rmse = 11.40114, val_rmse = 9.26007, ts_rmse = 9.24051, tr_c = 0.37386, val_c = 0.26151, ts_c = 0.25510
Epoch 103 time = 9.31, tr_rmse = 11.37935, val_rmse = 9.18178, ts_rmse = 9.15902, tr_c = 0.36167, val_c = 0.25182, ts_c = 0.24637
Epoch 104 time = 9.42, tr_rmse = 12.06568, val_rmse = 9.69470, ts_rmse = 9.62701, tr_c = 0.40496, val_c = 0.28561, ts_c = 0.27842
Epoch 105 time = 8.97, tr_rmse = 11.38487, val_rmse = 9.16275, ts_rmse = 9.13870, tr_c = 0.36898, val_c = 0.25790, ts_c = 0.25217
Epoch 106 time = 8.95, tr_rmse = 11.96571, val_rmse = 9.56119, ts_rmse = 9.50671, tr_c = 0.42529, val_c = 0.30187, ts_c = 0.29441
Epoch 107 time = 8.99, tr_rmse = 11.37544, val_rmse = 9.31393, ts_rmse = 9.28910, tr_c = 0.36529, val_c = 0.25436, ts_c = 0.24862
Epoch 108 time = 9.15, tr_rmse = 11.52029, val_rmse = 9.35353, ts_rmse = 9.34607, tr_c = 0.38205, val_c = 0.26826, ts_c = 0.26197
Epoch 109 time = 8.94, tr_rmse = 11.30600, val_rmse = 9.14069, ts_rmse = 9.13316, tr_c = 0.36022, val_c = 0.25138, ts_c = 0.24616
Epoch 110 time = 9.06, tr_rmse = 11.47536, val_rmse = 9.29977, ts_rmse = 9.27451, tr_c = 0.37103, val_c = 0.25891, ts_c = 0.25263
Epoch 111 time = 8.91, tr_rmse = 11.70083, val_rmse = 9.37088, ts_rmse = 9.35314, tr_c = 0.38409, val_c = 0.26882, ts_c = 0.26264
Epoch 112 time = 9.06, tr_rmse = 11.89644, val_rmse = 10.55736, ts_rmse = 10.56401, tr_c = 0.37530, val_c = 0.26246, ts_c = 0.25624
Epoch 113 time = 8.91, tr_rmse = 11.65831, val_rmse = 9.95303, ts_rmse = 9.97850, tr_c = 0.36747, val_c = 0.25671, ts_c = 0.25142
Epoch 114 time = 9.02, tr_rmse = 11.42944, val_rmse = 9.24083, ts_rmse = 9.24306, tr_c = 0.36245, val_c = 0.25297, ts_c = 0.24724
Epoch 115 time = 8.96, tr_rmse = 11.67072, val_rmse = 9.83944, ts_rmse = 9.89024, tr_c = 0.36477, val_c = 0.25332, ts_c = 0.24892
Epoch 116 time = 9.01, tr_rmse = 11.31010, val_rmse = 9.10097, ts_rmse = 9.11887, tr_c = 0.36285, val_c = 0.25425, ts_c = 0.24884
Epoch 117 time = 8.88, tr_rmse = 11.30528, val_rmse = 9.26420, ts_rmse = 9.24936, tr_c = 0.36830, val_c = 0.25885, ts_c = 0.25283
Epoch 118 time = 8.98, tr_rmse = 11.29169, val_rmse = 9.23864, ts_rmse = 9.22624, tr_c = 0.36202, val_c = 0.25260, ts_c = 0.24722
Epoch 119 time = 8.96, tr_rmse = 11.19792, val_rmse = 9.11351, ts_rmse = 9.12581, tr_c = 0.36346, val_c = 0.25645, ts_c = 0.25099
Epoch 120 time = 9.05, tr_rmse = 11.61180, val_rmse = 9.35633, ts_rmse = 9.32820, tr_c = 0.39586, val_c = 0.28063, ts_c = 0.27412
Epoch 121 time = 8.85, tr_rmse = 11.34085, val_rmse = 9.20514, ts_rmse = 9.20895, tr_c = 0.36078, val_c = 0.25167, ts_c = 0.24669
Epoch 122 time = 9.02, tr_rmse = 11.29047, val_rmse = 9.14608, ts_rmse = 9.16513, tr_c = 0.36771, val_c = 0.25857, ts_c = 0.25306
Epoch 123 time = 9.14, tr_rmse = 11.61002, val_rmse = 9.80235, ts_rmse = 9.83992, tr_c = 0.37919, val_c = 0.26803, ts_c = 0.26233
Epoch 124 time = 9.10, tr_rmse = 11.94388, val_rmse = 9.94394, ts_rmse = 9.92429, tr_c = 0.35916, val_c = 0.25098, ts_c = 0.24577
Epoch 125 time = 9.09, tr_rmse = 11.30471, val_rmse = 9.32061, ts_rmse = 9.33139, tr_c = 0.36377, val_c = 0.25459, ts_c = 0.24950
Epoch 126 time = 9.02, tr_rmse = 11.27412, val_rmse = 9.37997, ts_rmse = 9.37929, tr_c = 0.35909, val_c = 0.25066, ts_c = 0.24562
Epoch 127 time = 8.87, tr_rmse = 11.35638, val_rmse = 9.20837, ts_rmse = 9.22096, tr_c = 0.37244, val_c = 0.26249, ts_c = 0.25657
Epoch 128 time = 8.89, tr_rmse = 11.33776, val_rmse = 9.20102, ts_rmse = 9.20470, tr_c = 0.36849, val_c = 0.25849, ts_c = 0.25315
Epoch 129 time = 8.92, tr_rmse = 11.52539, val_rmse = 9.33517, ts_rmse = 9.33556, tr_c = 0.37776, val_c = 0.26779, ts_c = 0.26202
Epoch 130 time = 9.01, tr_rmse = 11.45616, val_rmse = 9.39308, ts_rmse = 9.38203, tr_c = 0.36078, val_c = 0.25351, ts_c = 0.24830
Epoch 131 time = 9.01, tr_rmse = 11.17754, val_rmse = 9.12070, ts_rmse = 9.12498, tr_c = 0.36034, val_c = 0.25270, ts_c = 0.24751
Epoch 132 time = 8.96, tr_rmse = 11.53374, val_rmse = 9.36793, ts_rmse = 9.34819, tr_c = 0.37497, val_c = 0.26222, ts_c = 0.25622
Epoch 133 time = 9.05, tr_rmse = 11.57543, val_rmse = 9.92123, ts_rmse = 9.96303, tr_c = 0.36785, val_c = 0.25838, ts_c = 0.25263
Epoch 134 time = 8.94, tr_rmse = 11.23877, val_rmse = 9.21431, ts_rmse = 9.21951, tr_c = 0.37181, val_c = 0.26376, ts_c = 0.25788
Epoch 135 time = 8.98, tr_rmse = 12.36886, val_rmse = 9.96180, ts_rmse = 9.92840, tr_c = 0.42180, val_c = 0.30769, ts_c = 0.30102
Epoch 136 time = 9.10, tr_rmse = 11.38600, val_rmse = 9.35586, ts_rmse = 9.35508, tr_c = 0.39397, val_c = 0.28111, ts_c = 0.27386
Epoch 137 time = 9.18, tr_rmse = 11.31410, val_rmse = 9.52963, ts_rmse = 9.56631, tr_c = 0.35855, val_c = 0.25098, ts_c = 0.24666
Epoch 138 time = 8.94, tr_rmse = 11.14519, val_rmse = 9.34718, ts_rmse = 9.36336, tr_c = 0.35784, val_c = 0.25204, ts_c = 0.24685
Epoch 139 time = 9.01, tr_rmse = 11.12508, val_rmse = 9.33751, ts_rmse = 9.33226, tr_c = 0.35660, val_c = 0.25019, ts_c = 0.24503
Epoch 140 time = 9.63, tr_rmse = 11.49631, val_rmse = 9.39116, ts_rmse = 9.39102, tr_c = 0.38357, val_c = 0.27194, ts_c = 0.26532
Epoch 141 time = 9.27, tr_rmse = 11.42483, val_rmse = 9.32793, ts_rmse = 9.34103, tr_c = 0.37873, val_c = 0.26836, ts_c = 0.26194
Epoch 142 time = 9.23, tr_rmse = 11.43525, val_rmse = 9.33651, ts_rmse = 9.32958, tr_c = 0.37426, val_c = 0.26295, ts_c = 0.25642
Epoch 143 time = 9.40, tr_rmse = 11.33231, val_rmse = 9.68819, ts_rmse = 9.70451, tr_c = 0.36120, val_c = 0.25471, ts_c = 0.24925
Epoch 144 time = 8.91, tr_rmse = 11.34209, val_rmse = 9.28790, ts_rmse = 9.30737, tr_c = 0.37553, val_c = 0.26474, ts_c = 0.25837
Epoch 145 time = 9.21, tr_rmse = 12.03746, val_rmse = 9.88929, ts_rmse = 9.83784, tr_c = 0.42851, val_c = 0.30425, ts_c = 0.29625
Epoch 146 time = 9.23, tr_rmse = 11.41999, val_rmse = 9.78663, ts_rmse = 9.81582, tr_c = 0.35853, val_c = 0.25177, ts_c = 0.24682
Epoch 147 time = 9.11, tr_rmse = 12.64206, val_rmse = 10.32783, ts_rmse = 10.25392, tr_c = 0.43413, val_c = 0.31748, ts_c = 0.31048
Epoch 148 time = 9.51, tr_rmse = 11.44831, val_rmse = 9.36571, ts_rmse = 9.37004, tr_c = 0.37885, val_c = 0.26883, ts_c = 0.26267
Epoch 149 time = 9.59, tr_rmse = 11.40456, val_rmse = 9.37309, ts_rmse = 9.36903, tr_c = 0.36183, val_c = 0.25381, ts_c = 0.24900
Epoch 150 time = 9.39, tr_rmse = 11.11085, val_rmse = 9.17626, ts_rmse = 9.16340, tr_c = 0.36395, val_c = 0.25679, ts_c = 0.25086
Epoch 151 time = 9.40, tr_rmse = 11.45920, val_rmse = 9.43215, ts_rmse = 9.44092, tr_c = 0.36874, val_c = 0.26032, ts_c = 0.25407
Epoch 152 time = 9.22, tr_rmse = 11.42822, val_rmse = 9.50313, ts_rmse = 9.49582, tr_c = 0.38134, val_c = 0.27075, ts_c = 0.26403
Epoch 153 time = 9.12, tr_rmse = 11.05820, val_rmse = 9.28374, ts_rmse = 9.28530, tr_c = 0.37500, val_c = 0.26792, ts_c = 0.26183
Epoch 154 time = 9.00, tr_rmse = 11.43391, val_rmse = 9.53927, ts_rmse = 9.50707, tr_c = 0.40442, val_c = 0.29028, ts_c = 0.28332
Epoch 155 time = 9.11, tr_rmse = 11.44204, val_rmse = 9.54678, ts_rmse = 9.52546, tr_c = 0.38213, val_c = 0.27410, ts_c = 0.26711
Epoch 156 time = 9.08, tr_rmse = 11.47264, val_rmse = 9.72309, ts_rmse = 9.72841, tr_c = 0.38168, val_c = 0.27423, ts_c = 0.26789
Epoch 157 time = 9.03, tr_rmse = 11.44218, val_rmse = 10.02846, ts_rmse = 10.02510, tr_c = 0.35694, val_c = 0.25126, ts_c = 0.24568
Epoch 158 time = 9.03, tr_rmse = 11.42072, val_rmse = 9.68191, ts_rmse = 9.70690, tr_c = 0.37982, val_c = 0.27165, ts_c = 0.26530
Epoch 159 time = 8.93, tr_rmse = 11.06860, val_rmse = 9.14078, ts_rmse = 9.15281, tr_c = 0.35760, val_c = 0.25334, ts_c = 0.24862
Epoch 160 time = 8.82, tr_rmse = 11.24876, val_rmse = 9.71125, ts_rmse = 9.71821, tr_c = 0.37526, val_c = 0.26597, ts_c = 0.25927
Epoch 161 time = 8.93, tr_rmse = 11.01326, val_rmse = 9.12827, ts_rmse = 9.13504, tr_c = 0.35921, val_c = 0.25477, ts_c = 0.24946
Epoch 162 time = 9.23, tr_rmse = 12.37618, val_rmse = 10.26922, ts_rmse = 10.15519, tr_c = 0.42397, val_c = 0.30419, ts_c = 0.29659
Epoch 163 time = 9.30, tr_rmse = 11.76375, val_rmse = 9.75821, ts_rmse = 9.75472, tr_c = 0.39192, val_c = 0.28217, ts_c = 0.27525
Epoch 164 time = 9.01, tr_rmse = 11.18651, val_rmse = 9.22398, ts_rmse = 9.23080, tr_c = 0.36294, val_c = 0.25434, ts_c = 0.24957
Epoch 165 time = 9.00, tr_rmse = 11.16190, val_rmse = 9.27273, ts_rmse = 9.26091, tr_c = 0.39063, val_c = 0.28182, ts_c = 0.27609
Epoch 166 time = 9.15, tr_rmse = 11.00196, val_rmse = 9.20589, ts_rmse = 9.22484, tr_c = 0.36348, val_c = 0.25939, ts_c = 0.25356
Epoch 167 time = 8.77, tr_rmse = 11.07943, val_rmse = 9.27690, ts_rmse = 9.25397, tr_c = 0.37129, val_c = 0.26483, ts_c = 0.25869
Epoch 168 time = 9.06, tr_rmse = 11.26095, val_rmse = 9.41331, ts_rmse = 9.39171, tr_c = 0.37467, val_c = 0.26621, ts_c = 0.25970
Epoch 169 time = 8.99, tr_rmse = 10.99721, val_rmse = 9.22775, ts_rmse = 9.24768, tr_c = 0.35620, val_c = 0.25261, ts_c = 0.24698
Epoch 170 time = 9.01, tr_rmse = 11.41140, val_rmse = 9.84292, ts_rmse = 9.88031, tr_c = 0.36613, val_c = 0.25944, ts_c = 0.25417
Epoch 171 time = 8.88, tr_rmse = 11.05844, val_rmse = 9.39746, ts_rmse = 9.42887, tr_c = 0.35742, val_c = 0.25261, ts_c = 0.24744
Epoch 172 time = 9.15, tr_rmse = 11.20609, val_rmse = 9.47374, ts_rmse = 9.50338, tr_c = 0.37751, val_c = 0.27086, ts_c = 0.26450
Epoch 173 time = 8.96, tr_rmse = 11.10970, val_rmse = 9.22246, ts_rmse = 9.24329, tr_c = 0.36239, val_c = 0.25830, ts_c = 0.25279
Epoch 174 time = 8.88, tr_rmse = 11.13557, val_rmse = 9.34774, ts_rmse = 9.31618, tr_c = 0.38103, val_c = 0.27215, ts_c = 0.26559
Epoch 175 time = 8.93, tr_rmse = 11.04277, val_rmse = 9.34381, ts_rmse = 9.34328, tr_c = 0.36223, val_c = 0.25785, ts_c = 0.25206
Epoch 176 time = 9.25, tr_rmse = 11.16678, val_rmse = 9.54608, ts_rmse = 9.55938, tr_c = 0.35777, val_c = 0.25272, ts_c = 0.24718
Epoch 177 time = 9.15, tr_rmse = 11.69894, val_rmse = 10.34202, ts_rmse = 10.39511, tr_c = 0.36692, val_c = 0.25819, ts_c = 0.25364
Epoch 178 time = 9.04, tr_rmse = 11.71480, val_rmse = 9.76083, ts_rmse = 9.74845, tr_c = 0.38936, val_c = 0.28058, ts_c = 0.27425
Epoch 179 time = 8.99, tr_rmse = 11.62095, val_rmse = 9.67208, ts_rmse = 9.67532, tr_c = 0.39661, val_c = 0.27916, ts_c = 0.27216
Epoch 180 time = 9.10, tr_rmse = 11.46090, val_rmse = 9.89378, ts_rmse = 9.93435, tr_c = 0.37755, val_c = 0.27187, ts_c = 0.26553
Epoch 181 time = 8.83, tr_rmse = 11.11012, val_rmse = 9.27158, ts_rmse = 9.27774, tr_c = 0.36173, val_c = 0.25606, ts_c = 0.25016
Epoch 182 time = 8.87, tr_rmse = 12.73362, val_rmse = 10.46947, ts_rmse = 10.39751, tr_c = 0.42864, val_c = 0.30987, ts_c = 0.30275
Epoch 183 time = 9.09, tr_rmse = 11.55728, val_rmse = 9.86438, ts_rmse = 9.83613, tr_c = 0.35574, val_c = 0.25247, ts_c = 0.24702
Epoch 184 time = 8.92, tr_rmse = 11.38268, val_rmse = 9.61568, ts_rmse = 9.59971, tr_c = 0.37299, val_c = 0.26759, ts_c = 0.26152
Epoch 185 time = 8.99, tr_rmse = 11.32633, val_rmse = 9.53522, ts_rmse = 9.53071, tr_c = 0.35518, val_c = 0.25126, ts_c = 0.24636
Epoch 186 time = 8.97, tr_rmse = 13.26600, val_rmse = 11.12548, ts_rmse = 10.96804, tr_c = 0.45445, val_c = 0.33644, ts_c = 0.32903
Epoch 187 time = 8.83, tr_rmse = 11.00835, val_rmse = 9.47154, ts_rmse = 9.48676, tr_c = 0.35448, val_c = 0.25165, ts_c = 0.24651
Epoch 188 time = 9.00, tr_rmse = 11.22604, val_rmse = 9.49174, ts_rmse = 9.50587, tr_c = 0.37700, val_c = 0.27122, ts_c = 0.26489
Epoch 189 time = 9.10, tr_rmse = 11.65760, val_rmse = 9.76303, ts_rmse = 9.78714, tr_c = 0.41169, val_c = 0.29483, ts_c = 0.28727
Epoch 190 time = 9.23, tr_rmse = 11.57839, val_rmse = 9.74334, ts_rmse = 9.73366, tr_c = 0.38293, val_c = 0.27520, ts_c = 0.26908
Epoch 191 time = 9.29, tr_rmse = 11.26498, val_rmse = 9.48013, ts_rmse = 9.50055, tr_c = 0.36522, val_c = 0.26021, ts_c = 0.25434
Epoch 192 time = 9.18, tr_rmse = 11.11865, val_rmse = 9.26745, ts_rmse = 9.26793, tr_c = 0.35793, val_c = 0.25544, ts_c = 0.25023
Epoch 193 time = 9.18, tr_rmse = 11.50938, val_rmse = 9.93278, ts_rmse = 9.93864, tr_c = 0.37548, val_c = 0.26699, ts_c = 0.26182
Epoch 194 time = 9.19, tr_rmse = 11.47673, val_rmse = 10.05181, ts_rmse = 10.03234, tr_c = 0.38050, val_c = 0.27026, ts_c = 0.26358
Epoch 195 time = 9.13, tr_rmse = 11.46414, val_rmse = 10.10611, ts_rmse = 10.09990, tr_c = 0.37735, val_c = 0.27194, ts_c = 0.26536
Epoch 196 time = 9.18, tr_rmse = 11.33487, val_rmse = 9.56770, ts_rmse = 9.61093, tr_c = 0.36670, val_c = 0.25764, ts_c = 0.25369
Epoch 197 time = 8.99, tr_rmse = 10.95802, val_rmse = 9.40345, ts_rmse = 9.43625, tr_c = 0.35322, val_c = 0.25047, ts_c = 0.24526
Epoch 198 time = 9.13, tr_rmse = 11.07573, val_rmse = 9.45267, ts_rmse = 9.43777, tr_c = 0.36189, val_c = 0.25618, ts_c = 0.25023
Epoch 199 time = 9.03, tr_rmse = 11.25885, val_rmse = 9.55649, ts_rmse = 9.53833, tr_c = 0.35376, val_c = 0.25180, ts_c = 0.24661
Epoch 200 time = 9.57, tr_rmse = 11.12383, val_rmse = 9.38186, ts_rmse = 9.40244, tr_c = 0.36134, val_c = 0.25862, ts_c = 0.25318
Epoch 201 time = 9.17, tr_rmse = 11.12646, val_rmse = 9.74152, ts_rmse = 9.77165, tr_c = 0.35477, val_c = 0.25273, ts_c = 0.24739
Epoch 202 time = 9.39, tr_rmse = 10.88519, val_rmse = 9.16364, ts_rmse = 9.16504, tr_c = 0.35758, val_c = 0.25569, ts_c = 0.25082
Epoch 203 time = 9.13, tr_rmse = 11.00628, val_rmse = 9.34643, ts_rmse = 9.31243, tr_c = 0.35659, val_c = 0.25515, ts_c = 0.24955
Epoch 204 time = 9.12, tr_rmse = 11.16019, val_rmse = 9.88437, ts_rmse = 9.86570, tr_c = 0.36322, val_c = 0.25879, ts_c = 0.25227
Epoch 205 time = 9.23, tr_rmse = 11.08986, val_rmse = 9.49834, ts_rmse = 9.51701, tr_c = 0.36598, val_c = 0.26200, ts_c = 0.25599
Epoch 206 time = 9.13, tr_rmse = 11.05721, val_rmse = 9.31117, ts_rmse = 9.32684, tr_c = 0.36559, val_c = 0.26152, ts_c = 0.25618
Epoch 207 time = 9.16, tr_rmse = 11.22754, val_rmse = 9.60482, ts_rmse = 9.59218, tr_c = 0.37816, val_c = 0.26970, ts_c = 0.26300
Epoch 208 time = 8.99, tr_rmse = 12.36802, val_rmse = 10.42602, ts_rmse = 10.35465, tr_c = 0.42480, val_c = 0.30943, ts_c = 0.30158
Epoch 209 time = 9.02, tr_rmse = 10.91718, val_rmse = 9.54652, ts_rmse = 9.53583, tr_c = 0.35216, val_c = 0.25086, ts_c = 0.24538
Epoch 210 time = 8.97, tr_rmse = 11.07220, val_rmse = 9.44090, ts_rmse = 9.45980, tr_c = 0.36215, val_c = 0.25822, ts_c = 0.25311
Epoch 211 time = 9.04, tr_rmse = 11.02114, val_rmse = 9.39564, ts_rmse = 9.39134, tr_c = 0.36139, val_c = 0.25881, ts_c = 0.25355
Epoch 212 time = 8.98, tr_rmse = 11.22438, val_rmse = 9.53068, ts_rmse = 9.49141, tr_c = 0.37611, val_c = 0.27159, ts_c = 0.26496
Epoch 213 time = 8.96, tr_rmse = 11.06414, val_rmse = 9.65843, ts_rmse = 9.68363, tr_c = 0.35846, val_c = 0.25656, ts_c = 0.25076
Epoch 214 time = 9.17, tr_rmse = 10.92939, val_rmse = 9.25858, ts_rmse = 9.24680, tr_c = 0.35735, val_c = 0.25721, ts_c = 0.25177
Epoch 215 time = 8.98, tr_rmse = 11.24747, val_rmse = 9.62981, ts_rmse = 9.64027, tr_c = 0.37332, val_c = 0.26694, ts_c = 0.26088
Epoch 216 time = 9.09, tr_rmse = 11.61408, val_rmse = 9.99181, ts_rmse = 9.93488, tr_c = 0.39401, val_c = 0.28640, ts_c = 0.27929
Epoch 217 time = 8.99, tr_rmse = 12.04590, val_rmse = 10.31526, ts_rmse = 10.25215, tr_c = 0.40257, val_c = 0.29002, ts_c = 0.28276
Epoch 218 time = 9.14, tr_rmse = 10.95620, val_rmse = 9.39259, ts_rmse = 9.39492, tr_c = 0.35940, val_c = 0.25691, ts_c = 0.25134
Epoch 219 time = 8.96, tr_rmse = 11.06372, val_rmse = 9.21410, ts_rmse = 9.22330, tr_c = 0.36033, val_c = 0.25552, ts_c = 0.25091
Epoch 220 time = 8.98, tr_rmse = 10.83487, val_rmse = 9.32950, ts_rmse = 9.33404, tr_c = 0.36780, val_c = 0.26412, ts_c = 0.25787
Epoch 221 time = 8.80, tr_rmse = 11.19125, val_rmse = 9.51938, ts_rmse = 9.52355, tr_c = 0.36445, val_c = 0.26120, ts_c = 0.25562
Epoch 222 time = 9.07, tr_rmse = 10.89673, val_rmse = 9.30874, ts_rmse = 9.34498, tr_c = 0.36244, val_c = 0.26101, ts_c = 0.25521
Epoch 223 time = 9.34, tr_rmse = 10.86890, val_rmse = 9.26432, ts_rmse = 9.25625, tr_c = 0.35400, val_c = 0.25266, ts_c = 0.24785
Epoch 224 time = 8.91, tr_rmse = 10.80099, val_rmse = 9.15529, ts_rmse = 9.16576, tr_c = 0.35137, val_c = 0.25037, ts_c = 0.24579
Epoch 225 time = 8.85, tr_rmse = 11.11140, val_rmse = 9.54163, ts_rmse = 9.53927, tr_c = 0.35783, val_c = 0.25795, ts_c = 0.25267
Epoch 226 time = 9.01, tr_rmse = 11.10636, val_rmse = 9.53211, ts_rmse = 9.56070, tr_c = 0.36612, val_c = 0.26238, ts_c = 0.25644
Epoch 227 time = 8.75, tr_rmse = 11.80725, val_rmse = 10.32145, ts_rmse = 10.29572, tr_c = 0.39013, val_c = 0.28028, ts_c = 0.27344
Epoch 228 time = 8.69, tr_rmse = 11.09903, val_rmse = 9.47346, ts_rmse = 9.43711, tr_c = 0.39163, val_c = 0.28488, ts_c = 0.27781
Epoch 229 time = 8.74, tr_rmse = 10.75149, val_rmse = 9.20312, ts_rmse = 9.20173, tr_c = 0.35122, val_c = 0.25096, ts_c = 0.24640
Epoch 230 time = 8.83, tr_rmse = 10.90631, val_rmse = 9.54786, ts_rmse = 9.49886, tr_c = 0.36468, val_c = 0.26150, ts_c = 0.25475
Epoch 231 time = 8.82, tr_rmse = 10.92225, val_rmse = 9.23875, ts_rmse = 9.24362, tr_c = 0.35353, val_c = 0.25225, ts_c = 0.24815
Epoch 232 time = 8.78, tr_rmse = 10.93569, val_rmse = 9.41811, ts_rmse = 9.39040, tr_c = 0.36023, val_c = 0.25888, ts_c = 0.25342
Epoch 233 time = 8.70, tr_rmse = 10.84509, val_rmse = 9.37019, ts_rmse = 9.37473, tr_c = 0.35714, val_c = 0.25694, ts_c = 0.25183
Epoch 234 time = 8.73, tr_rmse = 12.39860, val_rmse = 10.65704, ts_rmse = 10.55247, tr_c = 0.41933, val_c = 0.30509, ts_c = 0.29782
Epoch 235 time = 8.73, tr_rmse = 11.97927, val_rmse = 10.20529, ts_rmse = 10.17889, tr_c = 0.39984, val_c = 0.29342, ts_c = 0.28739
Epoch 236 time = 8.57, tr_rmse = 10.85678, val_rmse = 9.43420, ts_rmse = 9.44426, tr_c = 0.35451, val_c = 0.25307, ts_c = 0.24826
Epoch 237 time = 8.78, tr_rmse = 11.67046, val_rmse = 10.09832, ts_rmse = 10.06857, tr_c = 0.38443, val_c = 0.27580, ts_c = 0.26949
Epoch 238 time = 8.70, tr_rmse = 10.72606, val_rmse = 9.20977, ts_rmse = 9.20116, tr_c = 0.35201, val_c = 0.25295, ts_c = 0.24780
Epoch 239 time = 8.94, tr_rmse = 10.95755, val_rmse = 9.53616, ts_rmse = 9.56747, tr_c = 0.36009, val_c = 0.25735, ts_c = 0.25312
Epoch 240 time = 8.77, tr_rmse = 10.76626, val_rmse = 9.34247, ts_rmse = 9.31317, tr_c = 0.35513, val_c = 0.25563, ts_c = 0.25006
Epoch 241 time = 8.61, tr_rmse = 11.48687, val_rmse = 9.81899, ts_rmse = 9.76761, tr_c = 0.38347, val_c = 0.27518, ts_c = 0.26809
Epoch 242 time = 8.62, tr_rmse = 12.02850, val_rmse = 10.30740, ts_rmse = 10.24027, tr_c = 0.41176, val_c = 0.29783, ts_c = 0.29051
Epoch 243 time = 8.75, tr_rmse = 11.50523, val_rmse = 10.03899, ts_rmse = 10.05635, tr_c = 0.38559, val_c = 0.27946, ts_c = 0.27447
Epoch 244 time = 8.75, tr_rmse = 10.77822, val_rmse = 9.36046, ts_rmse = 9.34436, tr_c = 0.36338, val_c = 0.26462, ts_c = 0.25964
Epoch 245 time = 8.61, tr_rmse = 10.77898, val_rmse = 9.25678, ts_rmse = 9.23832, tr_c = 0.35215, val_c = 0.25253, ts_c = 0.24739
Epoch 246 time = 8.87, tr_rmse = 10.84766, val_rmse = 9.33920, ts_rmse = 9.32415, tr_c = 0.36543, val_c = 0.26482, ts_c = 0.25896
Epoch 247 time = 8.83, tr_rmse = 10.89226, val_rmse = 9.38863, ts_rmse = 9.34793, tr_c = 0.35624, val_c = 0.25740, ts_c = 0.25193
Epoch 248 time = 8.57, tr_rmse = 10.79261, val_rmse = 9.46731, ts_rmse = 9.47829, tr_c = 0.35048, val_c = 0.25095, ts_c = 0.24675
Epoch 249 time = 8.61, tr_rmse = 11.03798, val_rmse = 9.46671, ts_rmse = 9.41657, tr_c = 0.36450, val_c = 0.26450, ts_c = 0.25892
Epoch 250 time = 8.90, tr_rmse = 11.53505, val_rmse = 9.90949, ts_rmse = 9.81199, tr_c = 0.38696, val_c = 0.27831, ts_c = 0.27083
Epoch 251 time = 8.76, tr_rmse = 11.42634, val_rmse = 10.05545, ts_rmse = 10.05042, tr_c = 0.37682, val_c = 0.27110, ts_c = 0.26490
Epoch 252 time = 8.63, tr_rmse = 11.13601, val_rmse = 9.60814, ts_rmse = 9.57437, tr_c = 0.36991, val_c = 0.26707, ts_c = 0.26082
Epoch 253 time = 8.75, tr_rmse = 11.10732, val_rmse = 9.72272, ts_rmse = 9.74482, tr_c = 0.36607, val_c = 0.26368, ts_c = 0.25901
Epoch 254 time = 10.05, tr_rmse = 10.74062, val_rmse = 9.21721, ts_rmse = 9.22688, tr_c = 0.35043, val_c = 0.25167, ts_c = 0.24752
Epoch 255 time = 8.84, tr_rmse = 11.12758, val_rmse = 9.92905, ts_rmse = 9.96006, tr_c = 0.35554, val_c = 0.25481, ts_c = 0.25035
Epoch 256 time = 9.00, tr_rmse = 11.39279, val_rmse = 10.37080, ts_rmse = 10.40930, tr_c = 0.35344, val_c = 0.25239, ts_c = 0.24876
Epoch 257 time = 9.22, tr_rmse = 10.72966, val_rmse = 9.25421, ts_rmse = 9.22792, tr_c = 0.36211, val_c = 0.26338, ts_c = 0.25763
Epoch 258 time = 9.06, tr_rmse = 11.33248, val_rmse = 9.75671, ts_rmse = 9.71373, tr_c = 0.37582, val_c = 0.26903, ts_c = 0.26221
Epoch 259 time = 10.26, tr_rmse = 10.75576, val_rmse = 9.36539, ts_rmse = 9.40787, tr_c = 0.35333, val_c = 0.25269, ts_c = 0.24924
Epoch 260 time = 10.64, tr_rmse = 11.06786, val_rmse = 9.50234, ts_rmse = 9.48141, tr_c = 0.36918, val_c = 0.26642, ts_c = 0.26071
Epoch 261 time = 9.41, tr_rmse = 10.82685, val_rmse = 9.64724, ts_rmse = 9.66806, tr_c = 0.35886, val_c = 0.26086, ts_c = 0.25595
Epoch 262 time = 9.14, tr_rmse = 10.78920, val_rmse = 9.50731, ts_rmse = 9.53539, tr_c = 0.34867, val_c = 0.24977, ts_c = 0.24613
Epoch 263 time = 9.09, tr_rmse = 10.66263, val_rmse = 9.27510, ts_rmse = 9.25902, tr_c = 0.35311, val_c = 0.25443, ts_c = 0.24932
Epoch 264 time = 9.27, tr_rmse = 11.37250, val_rmse = 9.77787, ts_rmse = 9.69651, tr_c = 0.42179, val_c = 0.30996, ts_c = 0.30258
Epoch 265 time = 9.38, tr_rmse = 11.98619, val_rmse = 10.40380, ts_rmse = 10.30730, tr_c = 0.43657, val_c = 0.31728, ts_c = 0.30957
Epoch 266 time = 9.69, tr_rmse = 10.88139, val_rmse = 9.81540, ts_rmse = 9.79877, tr_c = 0.36132, val_c = 0.26145, ts_c = 0.25563
Epoch 267 time = 9.55, tr_rmse = 10.98192, val_rmse = 9.57513, ts_rmse = 9.53505, tr_c = 0.34740, val_c = 0.24957, ts_c = 0.24513
Epoch 268 time = 9.85, tr_rmse = 10.63223, val_rmse = 9.27149, ts_rmse = 9.26701, tr_c = 0.34855, val_c = 0.25128, ts_c = 0.24675
Epoch 269 time = 9.26, tr_rmse = 10.99660, val_rmse = 9.88710, ts_rmse = 9.84670, tr_c = 0.37453, val_c = 0.27285, ts_c = 0.26624
Epoch 270 time = 9.20, tr_rmse = 11.92757, val_rmse = 10.21691, ts_rmse = 10.14126, tr_c = 0.39800, val_c = 0.29146, ts_c = 0.28529
Epoch 271 time = 9.14, tr_rmse = 11.42148, val_rmse = 9.91293, ts_rmse = 9.89955, tr_c = 0.38191, val_c = 0.27658, ts_c = 0.27049
Epoch 272 time = 9.22, tr_rmse = 10.83131, val_rmse = 9.36621, ts_rmse = 9.35579, tr_c = 0.36438, val_c = 0.26556, ts_c = 0.26055
Epoch 273 time = 9.28, tr_rmse = 11.04573, val_rmse = 9.56890, ts_rmse = 9.52149, tr_c = 0.36435, val_c = 0.26345, ts_c = 0.25692
Epoch 274 time = 8.98, tr_rmse = 10.63124, val_rmse = 9.34323, ts_rmse = 9.32208, tr_c = 0.35502, val_c = 0.25652, ts_c = 0.25128
Epoch 275 time = 8.93, tr_rmse = 10.79848, val_rmse = 9.39405, ts_rmse = 9.37229, tr_c = 0.35111, val_c = 0.25223, ts_c = 0.24806
Epoch 276 time = 8.90, tr_rmse = 10.70592, val_rmse = 9.35860, ts_rmse = 9.33026, tr_c = 0.35381, val_c = 0.25653, ts_c = 0.25133
Epoch 277 time = 9.02, tr_rmse = 15.34340, val_rmse = 12.84106, ts_rmse = 12.63780, tr_c = 0.49159, val_c = 0.36493, ts_c = 0.35779
Epoch 278 time = 8.96, tr_rmse = 10.88945, val_rmse = 9.48851, ts_rmse = 9.42521, tr_c = 0.36102, val_c = 0.25987, ts_c = 0.25386
Epoch 279 time = 9.01, tr_rmse = 10.93226, val_rmse = 9.97163, ts_rmse = 9.95897, tr_c = 0.35760, val_c = 0.26039, ts_c = 0.25499
Epoch 280 time = 9.10, tr_rmse = 10.69714, val_rmse = 9.44940, ts_rmse = 9.38492, tr_c = 0.37207, val_c = 0.26949, ts_c = 0.26243
Epoch 281 time = 8.93, tr_rmse = 11.18545, val_rmse = 9.73839, ts_rmse = 9.65443, tr_c = 0.36260, val_c = 0.26286, ts_c = 0.25699
Epoch 282 time = 9.01, tr_rmse = 10.66493, val_rmse = 9.38275, ts_rmse = 9.36488, tr_c = 0.35366, val_c = 0.25716, ts_c = 0.25214
Epoch 283 time = 9.03, tr_rmse = 11.70137, val_rmse = 10.09228, ts_rmse = 9.98708, tr_c = 0.38774, val_c = 0.27758, ts_c = 0.27054
Epoch 284 time = 8.97, tr_rmse = 11.50819, val_rmse = 9.96909, ts_rmse = 9.90966, tr_c = 0.38372, val_c = 0.27891, ts_c = 0.27284
Epoch 285 time = 9.14, tr_rmse = 10.99422, val_rmse = 9.60786, ts_rmse = 9.54158, tr_c = 0.37329, val_c = 0.27203, ts_c = 0.26529
Epoch 286 time = 8.99, tr_rmse = 10.96400, val_rmse = 9.36886, ts_rmse = 9.34003, tr_c = 0.37598, val_c = 0.27634, ts_c = 0.27188
Epoch 287 time = 9.15, tr_rmse = 10.57509, val_rmse = 9.31044, ts_rmse = 9.27477, tr_c = 0.35190, val_c = 0.25667, ts_c = 0.25213
Epoch 288 time = 9.22, tr_rmse = 10.76937, val_rmse = 9.54112, ts_rmse = 9.52643, tr_c = 0.35767, val_c = 0.25714, ts_c = 0.25160
Epoch 289 time = 9.17, tr_rmse = 10.69757, val_rmse = 9.35367, ts_rmse = 9.30499, tr_c = 0.36530, val_c = 0.26713, ts_c = 0.26140
Epoch 290 time = 9.37, tr_rmse = 11.43468, val_rmse = 9.83580, ts_rmse = 9.77045, tr_c = 0.42331, val_c = 0.30965, ts_c = 0.30274
Epoch 291 time = 9.15, tr_rmse = 10.65551, val_rmse = 9.28813, ts_rmse = 9.28167, tr_c = 0.34740, val_c = 0.24925, ts_c = 0.24569
Epoch 292 time = 8.99, tr_rmse = 10.87266, val_rmse = 9.40124, ts_rmse = 9.34119, tr_c = 0.36609, val_c = 0.26768, ts_c = 0.26246
Epoch 293 time = 9.01, tr_rmse = 10.82106, val_rmse = 9.56674, ts_rmse = 9.49337, tr_c = 0.37092, val_c = 0.26892, ts_c = 0.26206
Epoch 294 time = 8.95, tr_rmse = 10.61560, val_rmse = 9.54193, ts_rmse = 9.55229, tr_c = 0.34714, val_c = 0.25148, ts_c = 0.24687
Epoch 295 time = 9.06, tr_rmse = 10.71376, val_rmse = 9.74866, ts_rmse = 9.73949, tr_c = 0.35757, val_c = 0.26212, ts_c = 0.25674
Epoch 296 time = 9.01, tr_rmse = 11.79932, val_rmse = 10.36303, ts_rmse = 10.24701, tr_c = 0.37361, val_c = 0.27095, ts_c = 0.26529
Epoch 297 time = 8.91, tr_rmse = 10.89245, val_rmse = 9.84488, ts_rmse = 9.84170, tr_c = 0.35979, val_c = 0.26345, ts_c = 0.25852
Epoch 298 time = 8.87, tr_rmse = 11.94690, val_rmse = 10.86750, ts_rmse = 10.84033, tr_c = 0.35068, val_c = 0.25527, ts_c = 0.25108
Epoch 299 time = 8.89, tr_rmse = 11.70850, val_rmse = 10.19858, ts_rmse = 10.08285, tr_c = 0.38878, val_c = 0.28468, ts_c = 0.27819
Epoch 300 time = 9.08, tr_rmse = 10.51913, val_rmse = 9.33776, ts_rmse = 9.30690, tr_c = 0.34893, val_c = 0.25391, ts_c = 0.24930
Epoch 301 time = 9.01, tr_rmse = 11.60899, val_rmse = 10.38102, ts_rmse = 10.31513, tr_c = 0.37367, val_c = 0.27375, ts_c = 0.26796
Epoch 302 time = 8.99, tr_rmse = 10.82809, val_rmse = 9.59563, ts_rmse = 9.56260, tr_c = 0.35194, val_c = 0.25428, ts_c = 0.25044
Epoch 303 time = 9.00, tr_rmse = 10.72089, val_rmse = 9.71160, ts_rmse = 9.70381, tr_c = 0.35422, val_c = 0.25447, ts_c = 0.24879
Epoch 304 time = 8.87, tr_rmse = 11.22049, val_rmse = 9.83275, ts_rmse = 9.73852, tr_c = 0.38285, val_c = 0.27746, ts_c = 0.27074
Epoch 305 time = 9.07, tr_rmse = 10.76724, val_rmse = 9.58930, ts_rmse = 9.57318, tr_c = 0.35240, val_c = 0.25377, ts_c = 0.24919
Epoch 306 time = 8.99, tr_rmse = 10.73278, val_rmse = 9.46826, ts_rmse = 9.45907, tr_c = 0.35554, val_c = 0.25996, ts_c = 0.25538
Epoch 307 time = 9.50, tr_rmse = 11.18396, val_rmse = 9.76112, ts_rmse = 9.69594, tr_c = 0.37317, val_c = 0.27067, ts_c = 0.26461
Epoch 308 time = 8.84, tr_rmse = 10.62409, val_rmse = 9.40131, ts_rmse = 9.36203, tr_c = 0.35366, val_c = 0.25714, ts_c = 0.25221
Epoch 309 time = 8.99, tr_rmse = 11.64272, val_rmse = 10.03538, ts_rmse = 9.84720, tr_c = 0.43123, val_c = 0.30595, ts_c = 0.29620
Epoch 310 time = 9.00, tr_rmse = 10.54351, val_rmse = 9.39218, ts_rmse = 9.41080, tr_c = 0.34843, val_c = 0.25279, ts_c = 0.24895
Epoch 311 time = 8.92, tr_rmse = 12.00794, val_rmse = 11.65963, ts_rmse = 11.71692, tr_c = 0.35520, val_c = 0.25518, ts_c = 0.25199
Epoch 312 time = 8.98, tr_rmse = 10.52628, val_rmse = 9.42982, ts_rmse = 9.45227, tr_c = 0.34619, val_c = 0.25049, ts_c = 0.24636
Epoch 313 time = 9.01, tr_rmse = 10.55662, val_rmse = 9.35918, ts_rmse = 9.34947, tr_c = 0.34579, val_c = 0.24920, ts_c = 0.24528
Epoch 314 time = 8.92, tr_rmse = 10.59668, val_rmse = 9.36936, ts_rmse = 9.38548, tr_c = 0.34773, val_c = 0.25013, ts_c = 0.24716
Epoch 315 time = 9.07, tr_rmse = 10.68762, val_rmse = 9.26749, ts_rmse = 9.23746, tr_c = 0.37141, val_c = 0.27242, ts_c = 0.26772
Epoch 316 time = 9.09, tr_rmse = 11.22098, val_rmse = 9.88873, ts_rmse = 9.89510, tr_c = 0.36381, val_c = 0.26298, ts_c = 0.25842
Epoch 317 time = 9.12, tr_rmse = 11.48000, val_rmse = 10.09590, ts_rmse = 10.02272, tr_c = 0.37537, val_c = 0.27244, ts_c = 0.26695
Epoch 318 time = 8.92, tr_rmse = 10.81258, val_rmse = 9.67203, ts_rmse = 9.64494, tr_c = 0.34970, val_c = 0.25220, ts_c = 0.24843
Epoch 319 time = 8.97, tr_rmse = 10.61457, val_rmse = 9.33330, ts_rmse = 9.29505, tr_c = 0.35023, val_c = 0.25495, ts_c = 0.25021
Epoch 320 time = 9.16, tr_rmse = 11.10853, val_rmse = 9.73009, ts_rmse = 9.71876, tr_c = 0.36510, val_c = 0.26491, ts_c = 0.25955
Epoch 321 time = 8.97, tr_rmse = 12.12476, val_rmse = 10.62866, ts_rmse = 10.34047, tr_c = 0.41509, val_c = 0.30120, ts_c = 0.29308
Epoch 322 time = 8.99, tr_rmse = 11.35118, val_rmse = 10.54904, ts_rmse = 10.48565, tr_c = 0.38252, val_c = 0.27428, ts_c = 0.26725
Epoch 323 time = 8.91, tr_rmse = 11.13491, val_rmse = 9.75374, ts_rmse = 9.62628, tr_c = 0.39826, val_c = 0.28873, ts_c = 0.28060
Epoch 324 time = 8.99, tr_rmse = 11.05099, val_rmse = 9.72438, ts_rmse = 9.69245, tr_c = 0.36713, val_c = 0.26718, ts_c = 0.26170
Epoch 325 time = 8.93, tr_rmse = 11.01047, val_rmse = 9.58858, ts_rmse = 9.59012, tr_c = 0.36447, val_c = 0.25963, ts_c = 0.25641
Epoch 326 time = 9.00, tr_rmse = 10.55429, val_rmse = 9.29878, ts_rmse = 9.24688, tr_c = 0.35821, val_c = 0.26187, ts_c = 0.25629
Epoch 327 time = 9.03, tr_rmse = 10.68737, val_rmse = 9.73120, ts_rmse = 9.78576, tr_c = 0.35055, val_c = 0.25357, ts_c = 0.24917
Epoch 328 time = 8.93, tr_rmse = 10.76261, val_rmse = 9.50154, ts_rmse = 9.50833, tr_c = 0.35262, val_c = 0.25260, ts_c = 0.24780
Epoch 329 time = 9.04, tr_rmse = 10.66531, val_rmse = 9.63131, ts_rmse = 9.65635, tr_c = 0.35200, val_c = 0.25558, ts_c = 0.25198
Epoch 330 time = 9.02, tr_rmse = 10.51987, val_rmse = 9.57071, ts_rmse = 9.56630, tr_c = 0.34606, val_c = 0.25133, ts_c = 0.24711
Epoch 331 time = 9.10, tr_rmse = 12.01164, val_rmse = 10.69146, ts_rmse = 10.64336, tr_c = 0.40507, val_c = 0.30337, ts_c = 0.29802
Epoch 332 time = 9.02, tr_rmse = 10.70089, val_rmse = 9.56285, ts_rmse = 9.55813, tr_c = 0.36773, val_c = 0.26668, ts_c = 0.26100
Epoch 333 time = 9.04, tr_rmse = 11.12620, val_rmse = 10.10694, ts_rmse = 10.07635, tr_c = 0.36572, val_c = 0.26644, ts_c = 0.26124
Epoch 334 time = 9.03, tr_rmse = 10.62846, val_rmse = 9.43308, ts_rmse = 9.38004, tr_c = 0.37258, val_c = 0.27690, ts_c = 0.27149
Epoch 335 time = 9.42, tr_rmse = 10.57817, val_rmse = 9.29780, ts_rmse = 9.25894, tr_c = 0.35067, val_c = 0.25576, ts_c = 0.25096
Epoch 336 time = 8.72, tr_rmse = 10.72040, val_rmse = 9.54357, ts_rmse = 9.57259, tr_c = 0.35534, val_c = 0.26007, ts_c = 0.25556
Epoch 337 time = 8.76, tr_rmse = 10.58077, val_rmse = 9.41761, ts_rmse = 9.45440, tr_c = 0.35131, val_c = 0.25607, ts_c = 0.25295
Epoch 338 time = 8.53, tr_rmse = 11.46695, val_rmse = 10.28777, ts_rmse = 10.22865, tr_c = 0.38444, val_c = 0.27639, ts_c = 0.27014
Epoch 339 time = 8.79, tr_rmse = 11.34796, val_rmse = 10.00801, ts_rmse = 9.89439, tr_c = 0.38259, val_c = 0.27900, ts_c = 0.27276
Epoch 340 time = 8.86, tr_rmse = 11.75216, val_rmse = 10.42904, ts_rmse = 10.33365, tr_c = 0.38099, val_c = 0.27660, ts_c = 0.27095
Epoch 341 time = 8.89, tr_rmse = 11.09715, val_rmse = 10.37528, ts_rmse = 10.36668, tr_c = 0.34352, val_c = 0.24798, ts_c = 0.24417
Epoch 342 time = 8.82, tr_rmse = 10.98776, val_rmse = 9.83481, ts_rmse = 9.82292, tr_c = 0.36264, val_c = 0.26510, ts_c = 0.26026
Epoch 343 time = 8.76, tr_rmse = 10.64211, val_rmse = 9.35164, ts_rmse = 9.30366, tr_c = 0.34472, val_c = 0.25092, ts_c = 0.24703
Epoch 344 time = 8.86, tr_rmse = 11.05254, val_rmse = 9.92913, ts_rmse = 9.89712, tr_c = 0.36774, val_c = 0.26810, ts_c = 0.26299
Epoch 345 time = 8.75, tr_rmse = 11.33654, val_rmse = 10.66524, ts_rmse = 10.71908, tr_c = 0.36249, val_c = 0.26240, ts_c = 0.25928
Epoch 346 time = 8.80, tr_rmse = 10.67953, val_rmse = 9.88124, ts_rmse = 9.92399, tr_c = 0.34281, val_c = 0.24888, ts_c = 0.24504
Epoch 347 time = 8.81, tr_rmse = 11.01666, val_rmse = 9.80621, ts_rmse = 9.79076, tr_c = 0.36175, val_c = 0.26217, ts_c = 0.25679
Epoch 348 time = 8.69, tr_rmse = 10.46829, val_rmse = 9.45033, ts_rmse = 9.44438, tr_c = 0.34527, val_c = 0.25155, ts_c = 0.24801
Epoch 349 time = 8.81, tr_rmse = 10.68020, val_rmse = 9.69637, ts_rmse = 9.61391, tr_c = 0.36423, val_c = 0.26334, ts_c = 0.25783
Epoch 350 time = 8.91, tr_rmse = 11.17869, val_rmse = 9.97388, ts_rmse = 9.98481, tr_c = 0.36530, val_c = 0.26484, ts_c = 0.25998
Epoch 351 time = 8.78, tr_rmse = 11.70925, val_rmse = 10.25893, ts_rmse = 10.12397, tr_c = 0.39765, val_c = 0.28359, ts_c = 0.27657
Epoch 352 time = 8.61, tr_rmse = 10.74268, val_rmse = 9.80792, ts_rmse = 9.81823, tr_c = 0.35882, val_c = 0.26142, ts_c = 0.25819
Epoch 353 time = 8.68, tr_rmse = 11.04693, val_rmse = 9.74124, ts_rmse = 9.66803, tr_c = 0.37096, val_c = 0.26949, ts_c = 0.26346
Epoch 354 time = 8.92, tr_rmse = 10.49994, val_rmse = 9.43416, ts_rmse = 9.45844, tr_c = 0.35021, val_c = 0.25496, ts_c = 0.25041
Epoch 355 time = 8.81, tr_rmse = 10.56126, val_rmse = 9.34483, ts_rmse = 9.27757, tr_c = 0.34630, val_c = 0.25195, ts_c = 0.24835
Epoch 356 time = 8.79, tr_rmse = 10.62368, val_rmse = 9.70609, ts_rmse = 9.72205, tr_c = 0.34355, val_c = 0.24894, ts_c = 0.24541
Epoch 357 time = 8.90, tr_rmse = 10.51389, val_rmse = 9.41370, ts_rmse = 9.39687, tr_c = 0.34412, val_c = 0.25040, ts_c = 0.24676
Epoch 358 time = 8.62, tr_rmse = 10.92298, val_rmse = 10.36549, ts_rmse = 10.36039, tr_c = 0.35524, val_c = 0.26114, ts_c = 0.25652
Epoch 359 time = 8.74, tr_rmse = 11.13074, val_rmse = 9.83178, ts_rmse = 9.79429, tr_c = 0.37109, val_c = 0.26796, ts_c = 0.26232
Epoch 360 time = 8.74, tr_rmse = 11.75357, val_rmse = 10.79355, ts_rmse = 10.74316, tr_c = 0.36099, val_c = 0.26619, ts_c = 0.26136
Epoch 361 time = 8.65, tr_rmse = 10.47948, val_rmse = 9.39465, ts_rmse = 9.39356, tr_c = 0.34289, val_c = 0.24912, ts_c = 0.24573
Epoch 362 time = 8.63, tr_rmse = 10.53845, val_rmse = 9.30307, ts_rmse = 9.28290, tr_c = 0.34646, val_c = 0.25324, ts_c = 0.24935
Epoch 363 time = 8.70, tr_rmse = 10.49256, val_rmse = 9.72709, ts_rmse = 9.79899, tr_c = 0.34568, val_c = 0.25182, ts_c = 0.24774
Epoch 364 time = 8.70, tr_rmse = 10.71602, val_rmse = 9.65488, ts_rmse = 9.66952, tr_c = 0.35301, val_c = 0.25754, ts_c = 0.25265
Epoch 365 time = 8.61, tr_rmse = 10.63490, val_rmse = 9.54828, ts_rmse = 9.48330, tr_c = 0.36436, val_c = 0.26909, ts_c = 0.26343
Epoch 366 time = 8.77, tr_rmse = 10.72190, val_rmse = 9.70255, ts_rmse = 9.74743, tr_c = 0.35585, val_c = 0.25645, ts_c = 0.25381
Epoch 367 time = 8.63, tr_rmse = 10.52494, val_rmse = 9.36156, ts_rmse = 9.33893, tr_c = 0.34612, val_c = 0.25318, ts_c = 0.24943
Epoch 368 time = 8.69, tr_rmse = 10.43859, val_rmse = 9.32384, ts_rmse = 9.28619, tr_c = 0.34222, val_c = 0.24955, ts_c = 0.24571
Epoch 369 time = 8.62, tr_rmse = 11.88649, val_rmse = 10.47440, ts_rmse = 10.29101, tr_c = 0.40927, val_c = 0.29170, ts_c = 0.28426
Epoch 370 time = 8.69, tr_rmse = 10.49447, val_rmse = 9.25957, ts_rmse = 9.22376, tr_c = 0.35599, val_c = 0.26136, ts_c = 0.25746
Epoch 371 time = 8.77, tr_rmse = 10.41139, val_rmse = 9.42468, ts_rmse = 9.44263, tr_c = 0.34296, val_c = 0.24938, ts_c = 0.24541
Epoch 372 time = 8.68, tr_rmse = 11.06350, val_rmse = 9.89036, ts_rmse = 9.78938, tr_c = 0.37430, val_c = 0.27249, ts_c = 0.26649
Epoch 373 time = 8.67, tr_rmse = 10.70699, val_rmse = 9.44461, ts_rmse = 9.38643, tr_c = 0.34533, val_c = 0.25080, ts_c = 0.24722
Epoch 374 time = 8.74, tr_rmse = 10.37240, val_rmse = 9.50769, ts_rmse = 9.52816, tr_c = 0.34660, val_c = 0.25389, ts_c = 0.24982
Epoch 375 time = 8.82, tr_rmse = 10.61222, val_rmse = 9.68997, ts_rmse = 9.70181, tr_c = 0.36064, val_c = 0.26069, ts_c = 0.25766
Epoch 376 time = 8.70, tr_rmse = 10.76937, val_rmse = 9.60601, ts_rmse = 9.59525, tr_c = 0.35082, val_c = 0.25395, ts_c = 0.25069
Epoch 377 time = 8.71, tr_rmse = 10.76611, val_rmse = 10.19362, ts_rmse = 10.25884, tr_c = 0.34689, val_c = 0.25400, ts_c = 0.25074
Epoch 378 time = 8.68, tr_rmse = 10.79494, val_rmse = 9.51866, ts_rmse = 9.47515, tr_c = 0.35261, val_c = 0.26015, ts_c = 0.25645
Epoch 379 time = 8.70, tr_rmse = 10.87275, val_rmse = 9.80290, ts_rmse = 9.75243, tr_c = 0.36439, val_c = 0.26375, ts_c = 0.25777
Epoch 380 time = 8.65, tr_rmse = 10.46373, val_rmse = 9.66712, ts_rmse = 9.71330, tr_c = 0.34451, val_c = 0.25061, ts_c = 0.24660
Epoch 381 time = 8.65, tr_rmse = 10.96421, val_rmse = 9.77178, ts_rmse = 9.73922, tr_c = 0.37843, val_c = 0.28084, ts_c = 0.27530
Epoch 382 time = 8.83, tr_rmse = 10.69730, val_rmse = 9.32054, ts_rmse = 9.26883, tr_c = 0.35682, val_c = 0.26184, ts_c = 0.25768
Epoch 383 time = 8.68, tr_rmse = 11.00253, val_rmse = 10.09263, ts_rmse = 10.12764, tr_c = 0.36278, val_c = 0.26521, ts_c = 0.26128
Epoch 384 time = 8.69, tr_rmse = 11.59918, val_rmse = 10.75090, ts_rmse = 10.64898, tr_c = 0.38085, val_c = 0.27650, ts_c = 0.27030
Epoch 385 time = 8.78, tr_rmse = 10.48764, val_rmse = 9.71883, ts_rmse = 9.72802, tr_c = 0.35541, val_c = 0.26065, ts_c = 0.25660
Epoch 386 time = 8.82, tr_rmse = 10.54629, val_rmse = 9.65487, ts_rmse = 9.69502, tr_c = 0.34749, val_c = 0.25362, ts_c = 0.24963
Epoch 387 time = 8.69, tr_rmse = 11.24678, val_rmse = 10.91197, ts_rmse = 10.98601, tr_c = 0.34803, val_c = 0.25277, ts_c = 0.24973
Epoch 388 time = 8.45, tr_rmse = 10.44382, val_rmse = 9.38014, ts_rmse = 9.37137, tr_c = 0.34565, val_c = 0.25352, ts_c = 0.24951
Epoch 389 time = 8.81, tr_rmse = 10.61667, val_rmse = 9.71095, ts_rmse = 9.66897, tr_c = 0.36048, val_c = 0.26422, ts_c = 0.25916
Epoch 390 time = 8.50, tr_rmse = 10.68529, val_rmse = 9.63229, ts_rmse = 9.62934, tr_c = 0.36487, val_c = 0.26702, ts_c = 0.26167
Epoch 391 time = 8.60, tr_rmse = 10.92731, val_rmse = 10.15724, ts_rmse = 10.12217, tr_c = 0.35713, val_c = 0.26172, ts_c = 0.25782
Epoch 392 time = 8.92, tr_rmse = 10.89143, val_rmse = 9.85734, ts_rmse = 9.80027, tr_c = 0.36128, val_c = 0.26289, ts_c = 0.25775
Epoch 393 time = 8.76, tr_rmse = 10.48427, val_rmse = 9.46832, ts_rmse = 9.45611, tr_c = 0.34992, val_c = 0.25683, ts_c = 0.25296
Epoch 394 time = 8.75, tr_rmse = 10.59814, val_rmse = 9.59276, ts_rmse = 9.53202, tr_c = 0.36200, val_c = 0.26428, ts_c = 0.25844
Epoch 395 time = 8.73, tr_rmse = 10.80394, val_rmse = 10.42670, ts_rmse = 10.44856, tr_c = 0.34287, val_c = 0.25142, ts_c = 0.24779
Epoch 396 time = 8.81, tr_rmse = 10.49983, val_rmse = 9.57585, ts_rmse = 9.48761, tr_c = 0.38513, val_c = 0.28772, ts_c = 0.28217
Epoch 397 time = 8.68, tr_rmse = 11.10584, val_rmse = 10.13626, ts_rmse = 10.04874, tr_c = 0.37013, val_c = 0.26960, ts_c = 0.26337
Epoch 398 time = 8.51, tr_rmse = 11.47386, val_rmse = 10.25153, ts_rmse = 10.15186, tr_c = 0.37899, val_c = 0.27741, ts_c = 0.27185
Epoch 399 time = 8.85, tr_rmse = 10.47513, val_rmse = 9.38093, ts_rmse = 9.37794, tr_c = 0.34676, val_c = 0.25114, ts_c = 0.24828
Epoch 400 time = 8.75, tr_rmse = 10.44474, val_rmse = 9.25356, ts_rmse = 9.18545, tr_c = 0.35637, val_c = 0.26179, ts_c = 0.25744
Epoch 401 time = 8.55, tr_rmse = 10.39209, val_rmse = 9.37329, ts_rmse = 9.45493, tr_c = 0.34547, val_c = 0.25229, ts_c = 0.24898
Epoch 402 time = 8.71, tr_rmse = 11.43476, val_rmse = 10.29420, ts_rmse = 10.10138, tr_c = 0.37828, val_c = 0.27499, ts_c = 0.26882
Epoch 403 time = 8.68, tr_rmse = 10.71388, val_rmse = 9.45984, ts_rmse = 9.40773, tr_c = 0.34584, val_c = 0.25359, ts_c = 0.24979
Epoch 404 time = 8.62, tr_rmse = 10.57744, val_rmse = 9.80744, ts_rmse = 9.80761, tr_c = 0.35086, val_c = 0.25802, ts_c = 0.25411
Epoch 405 time = 8.68, tr_rmse = 10.40124, val_rmse = 9.31375, ts_rmse = 9.22520, tr_c = 0.34380, val_c = 0.25307, ts_c = 0.24909
Epoch 406 time = 8.78, tr_rmse = 10.31561, val_rmse = 9.34742, ts_rmse = 9.36255, tr_c = 0.34083, val_c = 0.24919, ts_c = 0.24549
Epoch 407 time = 8.69, tr_rmse = 10.51223, val_rmse = 9.62904, ts_rmse = 9.61306, tr_c = 0.34753, val_c = 0.25495, ts_c = 0.25139
Epoch 408 time = 8.63, tr_rmse = 10.39865, val_rmse = 9.50912, ts_rmse = 9.49625, tr_c = 0.35750, val_c = 0.26076, ts_c = 0.25575
Epoch 409 time = 9.07, tr_rmse = 10.71932, val_rmse = 9.96456, ts_rmse = 10.02277, tr_c = 0.35283, val_c = 0.25935, ts_c = 0.25574
Epoch 410 time = 8.76, tr_rmse = 11.16451, val_rmse = 10.02067, ts_rmse = 9.97576, tr_c = 0.37745, val_c = 0.27820, ts_c = 0.27431
Epoch 411 time = 8.51, tr_rmse = 10.37173, val_rmse = 9.56178, ts_rmse = 9.52558, tr_c = 0.36160, val_c = 0.26695, ts_c = 0.26159
Epoch 412 time = 8.56, tr_rmse = 10.76422, val_rmse = 9.81259, ts_rmse = 9.81785, tr_c = 0.35533, val_c = 0.26008, ts_c = 0.25631
Epoch 413 time = 8.75, tr_rmse = 11.79925, val_rmse = 10.84235, ts_rmse = 10.74062, tr_c = 0.39335, val_c = 0.28531, ts_c = 0.27884
Epoch 414 time = 8.53, tr_rmse = 10.94096, val_rmse = 9.86545, ts_rmse = 9.89753, tr_c = 0.34726, val_c = 0.25092, ts_c = 0.24766
Epoch 415 time = 8.56, tr_rmse = 10.44067, val_rmse = 9.36036, ts_rmse = 9.39864, tr_c = 0.34724, val_c = 0.25486, ts_c = 0.25179
Epoch 416 time = 8.77, tr_rmse = 10.95412, val_rmse = 10.10700, ts_rmse = 10.16624, tr_c = 0.35324, val_c = 0.25595, ts_c = 0.25327
Epoch 417 time = 8.80, tr_rmse = 10.39774, val_rmse = 9.58510, ts_rmse = 9.60299, tr_c = 0.34118, val_c = 0.24904, ts_c = 0.24569
Epoch 418 time = 8.64, tr_rmse = 10.88318, val_rmse = 9.90045, ts_rmse = 9.93794, tr_c = 0.35999, val_c = 0.26399, ts_c = 0.26015
Epoch 419 time = 8.58, tr_rmse = 10.59559, val_rmse = 9.58872, ts_rmse = 9.51082, tr_c = 0.35020, val_c = 0.25847, ts_c = 0.25410
Epoch 420 time = 8.69, tr_rmse = 10.61444, val_rmse = 9.31619, ts_rmse = 9.28584, tr_c = 0.37572, val_c = 0.28226, ts_c = 0.27829
Epoch 421 time = 8.63, tr_rmse = 10.96847, val_rmse = 10.09287, ts_rmse = 10.09211, tr_c = 0.36124, val_c = 0.26566, ts_c = 0.26091
Epoch 422 time = 8.56, tr_rmse = 10.68784, val_rmse = 9.58646, ts_rmse = 9.61365, tr_c = 0.36288, val_c = 0.25978, ts_c = 0.25445
Epoch 423 time = 8.64, tr_rmse = 11.30561, val_rmse = 10.18467, ts_rmse = 10.13199, tr_c = 0.41344, val_c = 0.29696, ts_c = 0.28939
Epoch 424 time = 8.71, tr_rmse = 10.58705, val_rmse = 9.68178, ts_rmse = 9.68093, tr_c = 0.36975, val_c = 0.27014, ts_c = 0.26429
Epoch 425 time = 8.69, tr_rmse = 10.36266, val_rmse = 9.40459, ts_rmse = 9.38604, tr_c = 0.34014, val_c = 0.24976, ts_c = 0.24612
Epoch 426 time = 8.56, tr_rmse = 10.51887, val_rmse = 9.81594, ts_rmse = 9.81019, tr_c = 0.35745, val_c = 0.26297, ts_c = 0.25772
Epoch 427 time = 8.75, tr_rmse = 10.70898, val_rmse = 9.90201, ts_rmse = 9.91633, tr_c = 0.35509, val_c = 0.25783, ts_c = 0.25502
Epoch 428 time = 8.75, tr_rmse = 11.25894, val_rmse = 10.17042, ts_rmse = 10.10346, tr_c = 0.37478, val_c = 0.26874, ts_c = 0.26267
Epoch 429 time = 8.62, tr_rmse = 10.47342, val_rmse = 9.67917, ts_rmse = 9.74844, tr_c = 0.34941, val_c = 0.25691, ts_c = 0.25364
Epoch 430 time = 8.66, tr_rmse = 10.34825, val_rmse = 9.51759, ts_rmse = 9.51412, tr_c = 0.34322, val_c = 0.25167, ts_c = 0.24779
Epoch 431 time = 8.79, tr_rmse = 10.75354, val_rmse = 9.71709, ts_rmse = 9.66271, tr_c = 0.35999, val_c = 0.26496, ts_c = 0.25984
Epoch 432 time = 8.72, tr_rmse = 10.54881, val_rmse = 9.76690, ts_rmse = 9.73622, tr_c = 0.34694, val_c = 0.25216, ts_c = 0.24755
Epoch 433 time = 8.61, tr_rmse = 10.41717, val_rmse = 9.56881, ts_rmse = 9.53041, tr_c = 0.36033, val_c = 0.26224, ts_c = 0.25672
Epoch 434 time = 8.89, tr_rmse = 12.08548, val_rmse = 10.74787, ts_rmse = 10.62282, tr_c = 0.38975, val_c = 0.28220, ts_c = 0.27592
Epoch 435 time = 8.86, tr_rmse = 10.91068, val_rmse = 10.23132, ts_rmse = 10.23914, tr_c = 0.35611, val_c = 0.26172, ts_c = 0.25724
Epoch 436 time = 8.85, tr_rmse = 10.73428, val_rmse = 9.66632, ts_rmse = 9.57363, tr_c = 0.34778, val_c = 0.25645, ts_c = 0.25203
Epoch 437 time = 8.92, tr_rmse = 10.74128, val_rmse = 9.46212, ts_rmse = 9.43495, tr_c = 0.37429, val_c = 0.26817, ts_c = 0.26603
Epoch 438 time = 8.96, tr_rmse = 11.19299, val_rmse = 9.97123, ts_rmse = 9.83441, tr_c = 0.38507, val_c = 0.27627, ts_c = 0.26977
Epoch 439 time = 8.69, tr_rmse = 10.41726, val_rmse = 9.16358, ts_rmse = 9.17223, tr_c = 0.34310, val_c = 0.25012, ts_c = 0.24728
Epoch 440 time = 8.85, tr_rmse = 10.24223, val_rmse = 9.45809, ts_rmse = 9.45189, tr_c = 0.35009, val_c = 0.25719, ts_c = 0.25271
Epoch 441 time = 8.98, tr_rmse = 10.93714, val_rmse = 9.73955, ts_rmse = 9.69414, tr_c = 0.37639, val_c = 0.27566, ts_c = 0.26898
Epoch 442 time = 8.82, tr_rmse = 10.81220, val_rmse = 10.51729, ts_rmse = 10.62450, tr_c = 0.34290, val_c = 0.25174, ts_c = 0.24848
Epoch 443 time = 8.70, tr_rmse = 10.38010, val_rmse = 9.45343, ts_rmse = 9.49215, tr_c = 0.34236, val_c = 0.25006, ts_c = 0.24741
Epoch 444 time = 8.95, tr_rmse = 10.83724, val_rmse = 9.74444, ts_rmse = 9.69243, tr_c = 0.36402, val_c = 0.26408, ts_c = 0.25875
Epoch 445 time = 9.02, tr_rmse = 10.69011, val_rmse = 9.55758, ts_rmse = 9.57586, tr_c = 0.34444, val_c = 0.25403, ts_c = 0.25063
Epoch 446 time = 8.80, tr_rmse = 11.58515, val_rmse = 10.68352, ts_rmse = 10.58399, tr_c = 0.38736, val_c = 0.27925, ts_c = 0.27252
Epoch 447 time = 8.98, tr_rmse = 10.46481, val_rmse = 9.62381, ts_rmse = 9.65878, tr_c = 0.34818, val_c = 0.25528, ts_c = 0.25124
Epoch 448 time = 9.16, tr_rmse = 10.50363, val_rmse = 9.51039, ts_rmse = 9.45751, tr_c = 0.37793, val_c = 0.27545, ts_c = 0.26941
Epoch 449 time = 8.98, tr_rmse = 11.86379, val_rmse = 11.82849, ts_rmse = 11.94392, tr_c = 0.35404, val_c = 0.25900, ts_c = 0.25591
Epoch 450 time = 9.09, tr_rmse = 11.50014, val_rmse = 10.28464, ts_rmse = 10.18614, tr_c = 0.38466, val_c = 0.27842, ts_c = 0.27183
Epoch 451 time = 9.14, tr_rmse = 10.80707, val_rmse = 10.34001, ts_rmse = 10.40376, tr_c = 0.35426, val_c = 0.25872, ts_c = 0.25412
Epoch 452 time = 9.07, tr_rmse = 10.84248, val_rmse = 9.69200, ts_rmse = 9.60969, tr_c = 0.36012, val_c = 0.26082, ts_c = 0.25519
Epoch 453 time = 8.79, tr_rmse = 10.61146, val_rmse = 9.49827, ts_rmse = 9.44153, tr_c = 0.37615, val_c = 0.27658, ts_c = 0.27019
Epoch 454 time = 8.98, tr_rmse = 10.87800, val_rmse = 9.97346, ts_rmse = 9.87314, tr_c = 0.36099, val_c = 0.26621, ts_c = 0.26078
Epoch 455 time = 8.88, tr_rmse = 10.45322, val_rmse = 9.79031, ts_rmse = 9.82238, tr_c = 0.35307, val_c = 0.25912, ts_c = 0.25380
Epoch 456 time = 8.70, tr_rmse = 10.38279, val_rmse = 9.57980, ts_rmse = 9.67138, tr_c = 0.34304, val_c = 0.25219, ts_c = 0.24862
Epoch 457 time = 9.05, tr_rmse = 10.28377, val_rmse = 9.47896, ts_rmse = 9.56955, tr_c = 0.34135, val_c = 0.25104, ts_c = 0.24733
Epoch 458 time = 9.03, tr_rmse = 11.15291, val_rmse = 10.20622, ts_rmse = 10.19341, tr_c = 0.36757, val_c = 0.27175, ts_c = 0.26742
Epoch 459 time = 8.76, tr_rmse = 10.71524, val_rmse = 9.99740, ts_rmse = 9.96364, tr_c = 0.37219, val_c = 0.26976, ts_c = 0.26359
Epoch 460 time = 9.05, tr_rmse = 10.23716, val_rmse = 9.36692, ts_rmse = 9.38439, tr_c = 0.34507, val_c = 0.25424, ts_c = 0.24991
Epoch 461 time = 9.08, tr_rmse = 10.52579, val_rmse = 9.69226, ts_rmse = 9.72438, tr_c = 0.35399, val_c = 0.25999, ts_c = 0.25671
Epoch 462 time = 9.33, tr_rmse = 10.57016, val_rmse = 9.67025, ts_rmse = 9.73428, tr_c = 0.35520, val_c = 0.26188, ts_c = 0.25738
Epoch 463 time = 8.98, tr_rmse = 10.39072, val_rmse = 9.45762, ts_rmse = 9.46640, tr_c = 0.34022, val_c = 0.24863, ts_c = 0.24488
Epoch 464 time = 8.98, tr_rmse = 10.33939, val_rmse = 9.46207, ts_rmse = 9.53572, tr_c = 0.34117, val_c = 0.25046, ts_c = 0.24723
Epoch 465 time = 8.86, tr_rmse = 10.58601, val_rmse = 9.60566, ts_rmse = 9.60208, tr_c = 0.34722, val_c = 0.25538, ts_c = 0.25102
Epoch 466 time = 8.88, tr_rmse = 10.40368, val_rmse = 9.42322, ts_rmse = 9.49535, tr_c = 0.34442, val_c = 0.25435, ts_c = 0.25134
Epoch 467 time = 8.94, tr_rmse = 10.44912, val_rmse = 9.61474, ts_rmse = 9.56962, tr_c = 0.34484, val_c = 0.25358, ts_c = 0.24905
Epoch 468 time = 9.07, tr_rmse = 10.88601, val_rmse = 10.41060, ts_rmse = 10.46262, tr_c = 0.33938, val_c = 0.24862, ts_c = 0.24528
Epoch 469 time = 8.70, tr_rmse = 10.29965, val_rmse = 9.34710, ts_rmse = 9.33521, tr_c = 0.35154, val_c = 0.25986, ts_c = 0.25538
Epoch 470 time = 8.79, tr_rmse = 11.07366, val_rmse = 10.59813, ts_rmse = 10.69532, tr_c = 0.35722, val_c = 0.26204, ts_c = 0.25773
Epoch 471 time = 8.94, tr_rmse = 11.27228, val_rmse = 9.97668, ts_rmse = 9.89292, tr_c = 0.38340, val_c = 0.28374, ts_c = 0.27842
Epoch 472 time = 9.03, tr_rmse = 10.25298, val_rmse = 9.56180, ts_rmse = 9.55680, tr_c = 0.34916, val_c = 0.25848, ts_c = 0.25389
Epoch 473 time = 8.84, tr_rmse = 11.35453, val_rmse = 10.18647, ts_rmse = 10.08069, tr_c = 0.36805, val_c = 0.26848, ts_c = 0.26355
Epoch 474 time = 8.93, tr_rmse = 10.45006, val_rmse = 9.80129, ts_rmse = 9.82742, tr_c = 0.35663, val_c = 0.25832, ts_c = 0.25330
Epoch 475 time = 9.18, tr_rmse = 10.69941, val_rmse = 9.74997, ts_rmse = 9.64710, tr_c = 0.35582, val_c = 0.26281, ts_c = 0.25790
Epoch 476 time = 8.85, tr_rmse = 11.15392, val_rmse = 10.60907, ts_rmse = 10.58812, tr_c = 0.36062, val_c = 0.26423, ts_c = 0.26107
Epoch 477 time = 8.99, tr_rmse = 10.38602, val_rmse = 9.49573, ts_rmse = 9.55198, tr_c = 0.34919, val_c = 0.25636, ts_c = 0.25335
Epoch 478 time = 9.13, tr_rmse = 10.32740, val_rmse = 9.38761, ts_rmse = 9.41042, tr_c = 0.34351, val_c = 0.25327, ts_c = 0.24933
Epoch 479 time = 9.04, tr_rmse = 10.44405, val_rmse = 9.28921, ts_rmse = 9.32046, tr_c = 0.34429, val_c = 0.25325, ts_c = 0.24988
Epoch 480 time = 9.00, tr_rmse = 10.49145, val_rmse = 9.72836, ts_rmse = 9.79720, tr_c = 0.34693, val_c = 0.25387, ts_c = 0.24971
Epoch 481 time = 9.07, tr_rmse = 10.20557, val_rmse = 9.24819, ts_rmse = 9.26012, tr_c = 0.33874, val_c = 0.24954, ts_c = 0.24583
Epoch 482 time = 8.92, tr_rmse = 10.76154, val_rmse = 10.43528, ts_rmse = 10.46571, tr_c = 0.35221, val_c = 0.26295, ts_c = 0.25895
Epoch 483 time = 8.81, tr_rmse = 10.26711, val_rmse = 9.60495, ts_rmse = 9.59078, tr_c = 0.34675, val_c = 0.25497, ts_c = 0.25036
Epoch 484 time = 8.73, tr_rmse = 10.93331, val_rmse = 9.92195, ts_rmse = 9.87447, tr_c = 0.38784, val_c = 0.28591, ts_c = 0.27939
Epoch 485 time = 8.91, tr_rmse = 10.43321, val_rmse = 9.53648, ts_rmse = 9.51040, tr_c = 0.34161, val_c = 0.25129, ts_c = 0.24753
Epoch 486 time = 8.85, tr_rmse = 10.33321, val_rmse = 9.33024, ts_rmse = 9.29276, tr_c = 0.34321, val_c = 0.25322, ts_c = 0.24969
Epoch 487 time = 8.87, tr_rmse = 10.25392, val_rmse = 9.38982, ts_rmse = 9.47512, tr_c = 0.34146, val_c = 0.25034, ts_c = 0.24651
Epoch 488 time = 8.79, tr_rmse = 10.59532, val_rmse = 10.04363, ts_rmse = 10.09125, tr_c = 0.35129, val_c = 0.25789, ts_c = 0.25323
Epoch 489 time = 9.17, tr_rmse = 10.47576, val_rmse = 9.52931, ts_rmse = 9.49680, tr_c = 0.36185, val_c = 0.26881, ts_c = 0.26371
Epoch 490 time = 8.95, tr_rmse = 10.22557, val_rmse = 9.17169, ts_rmse = 9.19423, tr_c = 0.34447, val_c = 0.25431, ts_c = 0.25073
Epoch 491 time = 9.05, tr_rmse = 10.24527, val_rmse = 9.40272, ts_rmse = 9.41495, tr_c = 0.34041, val_c = 0.25036, ts_c = 0.24602
Epoch 492 time = 9.04, tr_rmse = 10.60030, val_rmse = 10.05839, ts_rmse = 10.09253, tr_c = 0.34082, val_c = 0.25226, ts_c = 0.24875
Epoch 493 time = 8.87, tr_rmse = 11.67945, val_rmse = 10.72347, ts_rmse = 10.69773, tr_c = 0.37637, val_c = 0.27739, ts_c = 0.27332
Epoch 494 time = 8.93, tr_rmse = 11.15799, val_rmse = 10.49596, ts_rmse = 10.58944, tr_c = 0.36106, val_c = 0.26533, ts_c = 0.26211
Epoch 495 time = 8.93, tr_rmse = 10.73830, val_rmse = 9.76281, ts_rmse = 9.68596, tr_c = 0.35186, val_c = 0.25756, ts_c = 0.25250
Epoch 496 time = 8.90, tr_rmse = 10.21389, val_rmse = 9.22230, ts_rmse = 9.23223, tr_c = 0.34349, val_c = 0.25363, ts_c = 0.24927
Epoch 497 time = 8.71, tr_rmse = 10.39454, val_rmse = 9.71266, ts_rmse = 9.73923, tr_c = 0.34983, val_c = 0.25672, ts_c = 0.25193
Epoch 498 time = 9.01, tr_rmse = 10.70300, val_rmse = 10.34836, ts_rmse = 10.38999, tr_c = 0.33957, val_c = 0.25067, ts_c = 0.24722
Epoch 499 time = 9.01, tr_rmse = 10.21581, val_rmse = 9.33685, ts_rmse = 9.41534, tr_c = 0.33881, val_c = 0.24838, ts_c = 0.24499
Epoch 500 time = 9.07, tr_rmse = 10.46447, val_rmse = 9.50358, ts_rmse = 9.51971, tr_c = 0.35620, val_c = 0.25908, ts_c = 0.25404
Epoch 501 time = 9.00, tr_rmse = 10.73285, val_rmse = 9.70103, ts_rmse = 9.65678, tr_c = 0.35914, val_c = 0.26488, ts_c = 0.25951
Epoch 502 time = 8.83, tr_rmse = 10.37594, val_rmse = 9.50391, ts_rmse = 9.45405, tr_c = 0.34866, val_c = 0.25887, ts_c = 0.25507
Epoch 503 time = 8.91, tr_rmse = 10.68245, val_rmse = 9.65408, ts_rmse = 9.56673, tr_c = 0.35973, val_c = 0.26782, ts_c = 0.26345
Epoch 504 time = 8.75, tr_rmse = 10.26672, val_rmse = 9.49260, ts_rmse = 9.54717, tr_c = 0.33841, val_c = 0.24879, ts_c = 0.24506
Epoch 505 time = 8.85, tr_rmse = 10.15527, val_rmse = 9.35712, ts_rmse = 9.36314, tr_c = 0.34419, val_c = 0.25503, ts_c = 0.25083
Epoch 506 time = 9.02, tr_rmse = 10.61646, val_rmse = 9.60294, ts_rmse = 9.59325, tr_c = 0.35566, val_c = 0.26056, ts_c = 0.25534
Epoch 507 time = 8.87, tr_rmse = 10.22923, val_rmse = 9.35908, ts_rmse = 9.36716, tr_c = 0.33717, val_c = 0.24842, ts_c = 0.24484
Epoch 508 time = 9.06, tr_rmse = 10.35460, val_rmse = 9.28266, ts_rmse = 9.32318, tr_c = 0.34111, val_c = 0.25087, ts_c = 0.24819
Epoch 509 time = 9.10, tr_rmse = 10.93165, val_rmse = 10.29248, ts_rmse = 10.35524, tr_c = 0.35570, val_c = 0.26056, ts_c = 0.25638
Epoch 510 time = 8.87, tr_rmse = 10.26140, val_rmse = 9.20853, ts_rmse = 9.23456, tr_c = 0.34010, val_c = 0.25103, ts_c = 0.24731
Epoch 511 time = 9.21, tr_rmse = 10.68775, val_rmse = 9.63479, ts_rmse = 9.66512, tr_c = 0.36487, val_c = 0.26618, ts_c = 0.26339
Epoch 512 time = 8.92, tr_rmse = 10.60538, val_rmse = 9.70447, ts_rmse = 9.67587, tr_c = 0.35628, val_c = 0.26102, ts_c = 0.25581
Epoch 513 time = 9.00, tr_rmse = 12.35381, val_rmse = 11.00081, ts_rmse = 10.91724, tr_c = 0.39598, val_c = 0.28730, ts_c = 0.28179
Epoch 514 time = 8.96, tr_rmse = 10.73231, val_rmse = 9.84959, ts_rmse = 9.81216, tr_c = 0.34687, val_c = 0.25504, ts_c = 0.25096
Epoch 515 time = 8.90, tr_rmse = 10.75244, val_rmse = 10.00275, ts_rmse = 10.05802, tr_c = 0.35626, val_c = 0.26291, ts_c = 0.25971
Epoch 516 time = 9.02, tr_rmse = 10.44488, val_rmse = 9.69772, ts_rmse = 9.61905, tr_c = 0.36314, val_c = 0.27351, ts_c = 0.26898
Epoch 517 time = 8.69, tr_rmse = 10.50575, val_rmse = 10.07349, ts_rmse = 10.13471, tr_c = 0.34399, val_c = 0.25450, ts_c = 0.25000
Epoch 518 time = 8.97, tr_rmse = 10.39543, val_rmse = 9.68895, ts_rmse = 9.77241, tr_c = 0.34391, val_c = 0.25277, ts_c = 0.25016
Epoch 519 time = 8.80, tr_rmse = 10.14465, val_rmse = 9.32682, ts_rmse = 9.39847, tr_c = 0.33842, val_c = 0.24949, ts_c = 0.24620
Epoch 520 time = 8.94, tr_rmse = 11.37354, val_rmse = 10.40323, ts_rmse = 10.34399, tr_c = 0.37173, val_c = 0.28031, ts_c = 0.27567
Epoch 521 time = 9.10, tr_rmse = 10.36059, val_rmse = 9.25949, ts_rmse = 9.18971, tr_c = 0.34711, val_c = 0.25649, ts_c = 0.25158
Epoch 522 time = 9.07, tr_rmse = 11.82352, val_rmse = 10.63055, ts_rmse = 10.38212, tr_c = 0.39153, val_c = 0.28916, ts_c = 0.28174
Epoch 523 time = 8.96, tr_rmse = 10.26787, val_rmse = 9.39476, ts_rmse = 9.48470, tr_c = 0.34228, val_c = 0.25245, ts_c = 0.24842
Epoch 524 time = 8.85, tr_rmse = 11.06926, val_rmse = 10.11578, ts_rmse = 10.07655, tr_c = 0.37576, val_c = 0.27023, ts_c = 0.26415
Epoch 525 time = 8.93, tr_rmse = 10.36778, val_rmse = 9.54965, ts_rmse = 9.51681, tr_c = 0.35949, val_c = 0.26640, ts_c = 0.26047
Epoch 526 time = 9.15, tr_rmse = 10.56871, val_rmse = 9.57970, ts_rmse = 9.61982, tr_c = 0.35832, val_c = 0.26564, ts_c = 0.26046
Epoch 527 time = 9.08, tr_rmse = 10.55979, val_rmse = 9.71105, ts_rmse = 9.77592, tr_c = 0.35286, val_c = 0.25871, ts_c = 0.25586
Epoch 528 time = 9.03, tr_rmse = 10.89747, val_rmse = 10.09285, ts_rmse = 10.03919, tr_c = 0.35318, val_c = 0.26263, ts_c = 0.25862
Epoch 529 time = 8.85, tr_rmse = 10.26520, val_rmse = 9.26305, ts_rmse = 9.25331, tr_c = 0.34049, val_c = 0.25118, ts_c = 0.24742
Epoch 530 time = 9.05, tr_rmse = 10.66659, val_rmse = 9.95517, ts_rmse = 10.06668, tr_c = 0.35434, val_c = 0.26009, ts_c = 0.25671
Epoch 531 time = 8.82, tr_rmse = 10.18610, val_rmse = 9.25695, ts_rmse = 9.22726, tr_c = 0.34338, val_c = 0.25314, ts_c = 0.24842
Epoch 532 time = 8.80, tr_rmse = 10.36495, val_rmse = 9.53621, ts_rmse = 9.63102, tr_c = 0.34540, val_c = 0.25576, ts_c = 0.25207
Epoch 533 time = 8.94, tr_rmse = 12.98388, val_rmse = 11.75860, ts_rmse = 11.60515, tr_c = 0.41027, val_c = 0.29670, ts_c = 0.29090
Epoch 534 time = 8.94, tr_rmse = 10.63888, val_rmse = 9.97671, ts_rmse = 10.03965, tr_c = 0.35014, val_c = 0.25741, ts_c = 0.25362
Epoch 535 time = 9.13, tr_rmse = 13.08658, val_rmse = 11.69186, ts_rmse = 11.51884, tr_c = 0.43100, val_c = 0.31549, ts_c = 0.30839
Epoch 536 time = 9.00, tr_rmse = 10.08640, val_rmse = 9.25329, ts_rmse = 9.26390, tr_c = 0.34424, val_c = 0.25615, ts_c = 0.25228
Epoch 537 time = 8.87, tr_rmse = 11.62315, val_rmse = 10.47413, ts_rmse = 10.42642, tr_c = 0.38027, val_c = 0.27877, ts_c = 0.27346
Epoch 538 time = 8.76, tr_rmse = 11.35222, val_rmse = 10.35723, ts_rmse = 10.30639, tr_c = 0.36021, val_c = 0.26564, ts_c = 0.26107
Epoch 539 time = 8.93, tr_rmse = 10.19582, val_rmse = 9.28089, ts_rmse = 9.27887, tr_c = 0.33665, val_c = 0.24853, ts_c = 0.24504
Epoch 540 time = 8.82, tr_rmse = 11.50901, val_rmse = 11.05230, ts_rmse = 11.11765, tr_c = 0.34950, val_c = 0.25426, ts_c = 0.25190
Epoch 541 time = 8.90, tr_rmse = 10.75347, val_rmse = 9.84356, ts_rmse = 9.82221, tr_c = 0.35278, val_c = 0.25860, ts_c = 0.25406
Epoch 542 time = 8.87, tr_rmse = 10.55212, val_rmse = 9.65913, ts_rmse = 9.71285, tr_c = 0.35002, val_c = 0.25759, ts_c = 0.25360
Epoch 543 time = 9.00, tr_rmse = 10.88816, val_rmse = 10.01530, ts_rmse = 9.99660, tr_c = 0.36511, val_c = 0.26931, ts_c = 0.26400
Epoch 544 time = 8.79, tr_rmse = 10.47928, val_rmse = 9.86485, ts_rmse = 9.93629, tr_c = 0.34020, val_c = 0.25004, ts_c = 0.24634
Epoch 545 time = 8.92, tr_rmse = 10.56858, val_rmse = 9.70668, ts_rmse = 9.80024, tr_c = 0.35904, val_c = 0.26372, ts_c = 0.26167
Epoch 546 time = 9.10, tr_rmse = 10.90502, val_rmse = 10.28621, ts_rmse = 10.30745, tr_c = 0.33705, val_c = 0.24956, ts_c = 0.24653
Epoch 547 time = 9.13, tr_rmse = 10.47475, val_rmse = 9.98543, ts_rmse = 10.03399, tr_c = 0.34338, val_c = 0.25332, ts_c = 0.24950
Epoch 548 time = 8.80, tr_rmse = 10.32576, val_rmse = 9.43448, ts_rmse = 9.42796, tr_c = 0.35775, val_c = 0.26842, ts_c = 0.26376
Epoch 549 time = 8.85, tr_rmse = 10.32283, val_rmse = 9.49256, ts_rmse = 9.56647, tr_c = 0.34517, val_c = 0.25326, ts_c = 0.24969
Epoch 550 time = 8.94, tr_rmse = 10.43533, val_rmse = 9.47559, ts_rmse = 9.54335, tr_c = 0.34907, val_c = 0.25624, ts_c = 0.25333
Epoch 551 time = 8.92, tr_rmse = 10.65101, val_rmse = 9.95581, ts_rmse = 10.03471, tr_c = 0.36115, val_c = 0.27101, ts_c = 0.26648
Epoch 552 time = 8.98, tr_rmse = 10.08485, val_rmse = 9.28684, ts_rmse = 9.35158, tr_c = 0.34004, val_c = 0.25225, ts_c = 0.24907
Epoch 553 time = 8.79, tr_rmse = 10.46815, val_rmse = 9.89120, ts_rmse = 9.93484, tr_c = 0.34486, val_c = 0.25498, ts_c = 0.25215
Epoch 554 time = 8.81, tr_rmse = 10.27884, val_rmse = 9.81822, ts_rmse = 9.87036, tr_c = 0.33962, val_c = 0.25198, ts_c = 0.24834
Epoch 555 time = 8.79, tr_rmse = 11.03516, val_rmse = 9.94045, ts_rmse = 9.89408, tr_c = 0.36920, val_c = 0.26695, ts_c = 0.26193
Epoch 556 time = 9.10, tr_rmse = 10.76554, val_rmse = 9.88380, ts_rmse = 9.84372, tr_c = 0.35167, val_c = 0.26081, ts_c = 0.25586
Epoch 557 time = 9.20, tr_rmse = 10.24015, val_rmse = 9.45366, ts_rmse = 9.56060, tr_c = 0.34395, val_c = 0.25555, ts_c = 0.25213
Epoch 558 time = 9.06, tr_rmse = 10.85371, val_rmse = 9.76684, ts_rmse = 9.63914, tr_c = 0.38887, val_c = 0.28443, ts_c = 0.27758
Epoch 559 time = 9.02, tr_rmse = 10.48922, val_rmse = 10.06210, ts_rmse = 10.15332, tr_c = 0.34437, val_c = 0.25420, ts_c = 0.25038
Epoch 560 time = 9.01, tr_rmse = 10.39561, val_rmse = 9.61939, ts_rmse = 9.75866, tr_c = 0.34459, val_c = 0.25577, ts_c = 0.25185
Epoch 561 time = 8.91, tr_rmse = 11.10457, val_rmse = 10.17039, ts_rmse = 10.04923, tr_c = 0.36290, val_c = 0.26929, ts_c = 0.26327
Epoch 562 time = 8.90, tr_rmse = 10.05416, val_rmse = 9.41967, ts_rmse = 9.48680, tr_c = 0.34340, val_c = 0.25563, ts_c = 0.25174
Epoch 563 time = 8.83, tr_rmse = 10.85711, val_rmse = 10.13493, ts_rmse = 10.17473, tr_c = 0.35521, val_c = 0.26415, ts_c = 0.25989
Epoch 564 time = 8.87, tr_rmse = 10.60726, val_rmse = 9.78519, ts_rmse = 9.65458, tr_c = 0.35300, val_c = 0.25858, ts_c = 0.25364
Epoch 565 time = 8.78, tr_rmse = 10.31908, val_rmse = 9.54962, ts_rmse = 9.57250, tr_c = 0.33740, val_c = 0.24931, ts_c = 0.24551
Epoch 566 time = 8.76, tr_rmse = 10.22046, val_rmse = 9.68302, ts_rmse = 9.79913, tr_c = 0.33771, val_c = 0.24968, ts_c = 0.24619
Epoch 567 time = 8.86, tr_rmse = 10.37264, val_rmse = 9.54658, ts_rmse = 9.58230, tr_c = 0.33903, val_c = 0.24899, ts_c = 0.24538
Epoch 568 time = 8.76, tr_rmse = 10.26843, val_rmse = 9.56269, ts_rmse = 9.63520, tr_c = 0.33726, val_c = 0.24974, ts_c = 0.24654
Epoch 569 time = 8.99, tr_rmse = 10.33962, val_rmse = 9.38378, ts_rmse = 9.41320, tr_c = 0.34190, val_c = 0.25418, ts_c = 0.25133
Epoch 570 time = 8.73, tr_rmse = 10.36502, val_rmse = 9.58211, ts_rmse = 9.58872, tr_c = 0.34621, val_c = 0.25762, ts_c = 0.25507
Epoch 571 time = 9.01, tr_rmse = 12.23469, val_rmse = 10.87260, ts_rmse = 10.65981, tr_c = 0.41466, val_c = 0.29530, ts_c = 0.28726
Epoch 572 time = 8.89, tr_rmse = 10.41831, val_rmse = 9.62226, ts_rmse = 9.65760, tr_c = 0.34525, val_c = 0.25578, ts_c = 0.25207
Epoch 573 time = 9.12, tr_rmse = 10.56785, val_rmse = 10.01347, ts_rmse = 9.97568, tr_c = 0.34909, val_c = 0.25680, ts_c = 0.25180
Epoch 574 time = 9.10, tr_rmse = 10.62764, val_rmse = 10.09964, ts_rmse = 10.13007, tr_c = 0.35325, val_c = 0.25766, ts_c = 0.25289
Epoch 575 time = 8.80, tr_rmse = 10.79579, val_rmse = 10.06713, ts_rmse = 10.05537, tr_c = 0.35351, val_c = 0.26367, ts_c = 0.26113
Epoch 576 time = 9.30, tr_rmse = 10.43968, val_rmse = 9.60276, ts_rmse = 9.61175, tr_c = 0.34694, val_c = 0.25818, ts_c = 0.25399
Epoch 577 time = 8.99, tr_rmse = 10.19836, val_rmse = 9.27967, ts_rmse = 9.27070, tr_c = 0.34505, val_c = 0.25393, ts_c = 0.24915
Epoch 578 time = 8.78, tr_rmse = 10.38359, val_rmse = 9.73194, ts_rmse = 9.84756, tr_c = 0.34758, val_c = 0.25590, ts_c = 0.25337
Epoch 579 time = 8.94, tr_rmse = 10.50116, val_rmse = 9.55260, ts_rmse = 9.61043, tr_c = 0.33906, val_c = 0.24839, ts_c = 0.24562
Epoch 580 time = 8.91, tr_rmse = 10.15152, val_rmse = 9.53629, ts_rmse = 9.57188, tr_c = 0.33602, val_c = 0.24916, ts_c = 0.24639
Epoch 581 time = 8.95, tr_rmse = 10.56130, val_rmse = 9.81295, ts_rmse = 9.77024, tr_c = 0.35351, val_c = 0.26107, ts_c = 0.25607
Epoch 582 time = 8.85, tr_rmse = 10.27851, val_rmse = 9.38097, ts_rmse = 9.45282, tr_c = 0.34541, val_c = 0.25395, ts_c = 0.25182
Epoch 583 time = 8.87, tr_rmse = 10.57844, val_rmse = 9.90092, ts_rmse = 10.02776, tr_c = 0.34723, val_c = 0.25691, ts_c = 0.25383
Epoch 584 time = 9.20, tr_rmse = 10.07181, val_rmse = 9.31513, ts_rmse = 9.31849, tr_c = 0.35135, val_c = 0.26199, ts_c = 0.25772
Epoch 585 time = 9.02, tr_rmse = 10.68408, val_rmse = 9.89055, ts_rmse = 9.87358, tr_c = 0.36125, val_c = 0.26673, ts_c = 0.26161
Epoch 586 time = 9.07, tr_rmse = 10.43862, val_rmse = 9.89626, ts_rmse = 10.06335, tr_c = 0.34265, val_c = 0.25392, ts_c = 0.25094
Epoch 587 time = 9.06, tr_rmse = 10.32072, val_rmse = 9.51291, ts_rmse = 9.57656, tr_c = 0.35258, val_c = 0.26091, ts_c = 0.25833
Epoch 588 time = 8.91, tr_rmse = 10.59630, val_rmse = 9.78585, ts_rmse = 9.92019, tr_c = 0.36059, val_c = 0.26513, ts_c = 0.26305
Epoch 589 time = 9.31, tr_rmse = 10.20288, val_rmse = 9.26035, ts_rmse = 9.23995, tr_c = 0.35196, val_c = 0.26183, ts_c = 0.25718
Epoch 590 time = 8.89, tr_rmse = 11.71393, val_rmse = 10.99213, ts_rmse = 11.06103, tr_c = 0.37609, val_c = 0.27958, ts_c = 0.27670
Epoch 591 time = 9.10, tr_rmse = 11.00957, val_rmse = 10.61758, ts_rmse = 10.66596, tr_c = 0.35084, val_c = 0.26133, ts_c = 0.25742
Epoch 592 time = 8.83, tr_rmse = 10.04295, val_rmse = 9.42543, ts_rmse = 9.39945, tr_c = 0.34485, val_c = 0.25719, ts_c = 0.25271
Epoch 593 time = 9.04, tr_rmse = 10.55734, val_rmse = 9.75049, ts_rmse = 9.66470, tr_c = 0.36543, val_c = 0.27291, ts_c = 0.26742
Epoch 594 time = 8.97, tr_rmse = 10.04728, val_rmse = 9.25414, ts_rmse = 9.26612, tr_c = 0.33676, val_c = 0.25074, ts_c = 0.24747
Epoch 595 time = 9.03, tr_rmse = 10.64891, val_rmse = 9.80154, ts_rmse = 9.77641, tr_c = 0.35798, val_c = 0.26215, ts_c = 0.25735
Epoch 596 time = 9.11, tr_rmse = 10.88980, val_rmse = 9.83907, ts_rmse = 9.86244, tr_c = 0.36105, val_c = 0.26874, ts_c = 0.26373
Epoch 597 time = 9.04, tr_rmse = 10.25640, val_rmse = 9.44942, ts_rmse = 9.47645, tr_c = 0.33964, val_c = 0.25247, ts_c = 0.24960
Epoch 598 time = 8.68, tr_rmse = 10.43064, val_rmse = 9.42337, ts_rmse = 9.51283, tr_c = 0.34251, val_c = 0.25196, ts_c = 0.24994
Epoch 599 time = 8.98, tr_rmse = 10.27517, val_rmse = 9.33628, ts_rmse = 9.34628, tr_c = 0.34142, val_c = 0.24967, ts_c = 0.24561
Epoch 600 time = 8.92, tr_rmse = 10.17221, val_rmse = 9.62410, ts_rmse = 9.67239, tr_c = 0.33846, val_c = 0.25169, ts_c = 0.24865
Epoch 601 time = 8.90, tr_rmse = 10.11414, val_rmse = 9.26462, ts_rmse = 9.31352, tr_c = 0.33849, val_c = 0.24990, ts_c = 0.24756
Epoch 602 time = 8.83, tr_rmse = 10.21221, val_rmse = 9.57745, ts_rmse = 9.64470, tr_c = 0.34224, val_c = 0.25307, ts_c = 0.24914
Epoch 603 time = 9.45, tr_rmse = 10.63426, val_rmse = 9.69665, ts_rmse = 9.69132, tr_c = 0.35585, val_c = 0.26478, ts_c = 0.25998
Epoch 604 time = 9.19, tr_rmse = 10.88800, val_rmse = 10.13372, ts_rmse = 10.09121, tr_c = 0.36575, val_c = 0.26763, ts_c = 0.26151
Epoch 605 time = 8.83, tr_rmse = 11.09139, val_rmse = 10.07822, ts_rmse = 9.99219, tr_c = 0.37836, val_c = 0.27770, ts_c = 0.27131
Epoch 606 time = 8.96, tr_rmse = 10.28261, val_rmse = 9.50893, ts_rmse = 9.49706, tr_c = 0.35195, val_c = 0.26143, ts_c = 0.25648
Epoch 607 time = 9.00, tr_rmse = 10.57915, val_rmse = 9.39615, ts_rmse = 9.34270, tr_c = 0.35418, val_c = 0.26221, ts_c = 0.25797
Epoch 608 time = 8.91, tr_rmse = 10.10584, val_rmse = 9.31262, ts_rmse = 9.32520, tr_c = 0.33672, val_c = 0.25060, ts_c = 0.24709
Epoch 609 time = 8.85, tr_rmse = 10.35303, val_rmse = 9.66649, ts_rmse = 9.73497, tr_c = 0.34196, val_c = 0.25301, ts_c = 0.25067
Epoch 610 time = 8.80, tr_rmse = 10.29074, val_rmse = 9.49629, ts_rmse = 9.54730, tr_c = 0.34075, val_c = 0.25427, ts_c = 0.25086
Epoch 611 time = 9.23, tr_rmse = 10.17635, val_rmse = 9.52017, ts_rmse = 9.60534, tr_c = 0.34342, val_c = 0.25706, ts_c = 0.25394
Epoch 612 time = 8.66, tr_rmse = 10.07883, val_rmse = 9.27748, ts_rmse = 9.31422, tr_c = 0.33589, val_c = 0.24964, ts_c = 0.24673
Epoch 613 time = 9.06, tr_rmse = 11.11596, val_rmse = 10.33640, ts_rmse = 10.38836, tr_c = 0.37450, val_c = 0.27595, ts_c = 0.27372
Epoch 614 time = 9.06, tr_rmse = 10.22184, val_rmse = 9.45067, ts_rmse = 9.48792, tr_c = 0.33817, val_c = 0.25000, ts_c = 0.24683
Epoch 615 time = 8.82, tr_rmse = 10.53739, val_rmse = 9.66525, ts_rmse = 9.68842, tr_c = 0.35412, val_c = 0.26152, ts_c = 0.25668
Epoch 616 time = 9.04, tr_rmse = 10.78926, val_rmse = 9.86718, ts_rmse = 9.83909, tr_c = 0.35424, val_c = 0.26452, ts_c = 0.26045
Epoch 617 time = 8.84, tr_rmse = 10.49465, val_rmse = 9.65189, ts_rmse = 9.80872, tr_c = 0.34647, val_c = 0.25789, ts_c = 0.25498
Epoch 618 time = 9.04, tr_rmse = 10.17556, val_rmse = 9.52335, ts_rmse = 9.55289, tr_c = 0.34207, val_c = 0.25422, ts_c = 0.25037
Epoch 619 time = 8.66, tr_rmse = 10.28663, val_rmse = 9.89188, ts_rmse = 9.97698, tr_c = 0.33808, val_c = 0.24990, ts_c = 0.24680
Epoch 620 time = 9.01, tr_rmse = 10.42686, val_rmse = 9.43764, ts_rmse = 9.46753, tr_c = 0.33984, val_c = 0.25206, ts_c = 0.24891
Epoch 621 time = 8.77, tr_rmse = 10.48077, val_rmse = 9.92808, ts_rmse = 9.95372, tr_c = 0.34440, val_c = 0.25385, ts_c = 0.25064
Epoch 622 time = 8.65, tr_rmse = 11.29342, val_rmse = 11.21263, ts_rmse = 11.30607, tr_c = 0.34610, val_c = 0.25555, ts_c = 0.25318
Epoch 623 time = 8.79, tr_rmse = 10.55100, val_rmse = 9.90397, ts_rmse = 9.84575, tr_c = 0.35714, val_c = 0.26166, ts_c = 0.25643
Epoch 624 time = 8.81, tr_rmse = 10.05047, val_rmse = 9.32021, ts_rmse = 9.33497, tr_c = 0.33713, val_c = 0.25001, ts_c = 0.24638
Epoch 625 time = 9.07, tr_rmse = 10.30379, val_rmse = 9.21623, ts_rmse = 9.21286, tr_c = 0.34642, val_c = 0.25830, ts_c = 0.25512
Epoch 626 time = 9.02, tr_rmse = 10.36675, val_rmse = 9.99580, ts_rmse = 10.10517, tr_c = 0.33917, val_c = 0.25165, ts_c = 0.24813
Epoch 627 time = 9.10, tr_rmse = 10.09923, val_rmse = 9.41415, ts_rmse = 9.46575, tr_c = 0.33536, val_c = 0.24844, ts_c = 0.24532
Epoch 628 time = 9.13, tr_rmse = 10.43371, val_rmse = 10.11805, ts_rmse = 10.22859, tr_c = 0.33722, val_c = 0.24992, ts_c = 0.24803
Epoch 629 time = 8.95, tr_rmse = 10.13485, val_rmse = 9.27635, ts_rmse = 9.30452, tr_c = 0.33528, val_c = 0.24846, ts_c = 0.24572
Epoch 630 time = 9.08, tr_rmse = 10.36447, val_rmse = 9.62466, ts_rmse = 9.52463, tr_c = 0.37441, val_c = 0.28059, ts_c = 0.27449
Epoch 631 time = 9.02, tr_rmse = 10.21897, val_rmse = 9.29577, ts_rmse = 9.33827, tr_c = 0.34280, val_c = 0.25607, ts_c = 0.25315
Epoch 632 time = 8.74, tr_rmse = 10.08844, val_rmse = 9.21507, ts_rmse = 9.19727, tr_c = 0.35651, val_c = 0.26841, ts_c = 0.26447
Epoch 633 time = 8.91, tr_rmse = 10.14871, val_rmse = 9.60102, ts_rmse = 9.61324, tr_c = 0.33784, val_c = 0.25169, ts_c = 0.24810
Epoch 634 time = 8.84, tr_rmse = 10.02308, val_rmse = 9.27964, ts_rmse = 9.30099, tr_c = 0.34834, val_c = 0.26256, ts_c = 0.25881
Epoch 635 time = 8.91, tr_rmse = 10.20960, val_rmse = 9.50429, ts_rmse = 9.44673, tr_c = 0.34242, val_c = 0.25543, ts_c = 0.25157
Epoch 636 time = 8.84, tr_rmse = 10.14837, val_rmse = 9.44515, ts_rmse = 9.42587, tr_c = 0.34017, val_c = 0.25160, ts_c = 0.24746
Epoch 637 time = 8.91, tr_rmse = 10.37604, val_rmse = 9.52682, ts_rmse = 9.59820, tr_c = 0.34152, val_c = 0.25504, ts_c = 0.25220
Epoch 638 time = 9.05, tr_rmse = 11.20807, val_rmse = 10.17822, ts_rmse = 10.15172, tr_c = 0.37429, val_c = 0.27460, ts_c = 0.26863
Epoch 639 time = 8.61, tr_rmse = 10.39423, val_rmse = 9.65654, ts_rmse = 9.58671, tr_c = 0.34308, val_c = 0.25495, ts_c = 0.25057
Epoch 640 time = 8.79, tr_rmse = 10.09450, val_rmse = 9.32431, ts_rmse = 9.32696, tr_c = 0.34455, val_c = 0.25556, ts_c = 0.25101
Epoch 641 time = 9.00, tr_rmse = 10.92642, val_rmse = 9.33793, ts_rmse = 9.38639, tr_c = 0.35746, val_c = 0.26201, ts_c = 0.25988
Epoch 642 time = 9.10, tr_rmse = 10.49871, val_rmse = 9.36468, ts_rmse = 9.35599, tr_c = 0.34072, val_c = 0.25093, ts_c = 0.24861
Epoch 643 time = 8.91, tr_rmse = 10.10160, val_rmse = 9.24138, ts_rmse = 9.21214, tr_c = 0.34786, val_c = 0.25676, ts_c = 0.25180
Epoch 644 time = 9.02, tr_rmse = 10.26332, val_rmse = 9.61778, ts_rmse = 9.55897, tr_c = 0.34611, val_c = 0.25816, ts_c = 0.25363
Epoch 645 time = 8.93, tr_rmse = 10.26889, val_rmse = 9.76974, ts_rmse = 9.78909, tr_c = 0.34252, val_c = 0.25538, ts_c = 0.25142
Epoch 646 time = 9.06, tr_rmse = 11.06919, val_rmse = 10.31878, ts_rmse = 10.23267, tr_c = 0.35993, val_c = 0.27127, ts_c = 0.26658
Epoch 647 time = 9.02, tr_rmse = 10.28764, val_rmse = 9.74724, ts_rmse = 9.86179, tr_c = 0.33883, val_c = 0.25220, ts_c = 0.24922
Epoch 648 time = 8.85, tr_rmse = 10.03931, val_rmse = 9.38874, ts_rmse = 9.34683, tr_c = 0.34409, val_c = 0.25710, ts_c = 0.25274
Epoch 649 time = 8.62, tr_rmse = 10.60917, val_rmse = 9.83361, ts_rmse = 9.83044, tr_c = 0.35499, val_c = 0.26475, ts_c = 0.25979
Epoch 650 time = 9.00, tr_rmse = 11.05394, val_rmse = 10.11738, ts_rmse = 10.01350, tr_c = 0.36974, val_c = 0.27300, ts_c = 0.26732
Epoch 651 time = 9.01, tr_rmse = 10.46156, val_rmse = 10.30316, ts_rmse = 10.33027, tr_c = 0.33753, val_c = 0.25219, ts_c = 0.24878
Epoch 652 time = 8.76, tr_rmse = 10.23021, val_rmse = 9.73780, ts_rmse = 9.81645, tr_c = 0.33987, val_c = 0.25329, ts_c = 0.25006
Epoch 653 time = 8.77, tr_rmse = 10.24237, val_rmse = 9.55637, ts_rmse = 9.47341, tr_c = 0.34490, val_c = 0.25856, ts_c = 0.25440
Epoch 654 time = 8.93, tr_rmse = 10.36634, val_rmse = 9.34088, ts_rmse = 9.27760, tr_c = 0.34981, val_c = 0.26192, ts_c = 0.25781
Epoch 655 time = 9.03, tr_rmse = 10.32846, val_rmse = 9.17641, ts_rmse = 9.21860, tr_c = 0.34621, val_c = 0.25725, ts_c = 0.25520
Epoch 656 time = 9.02, tr_rmse = 10.88701, val_rmse = 10.25234, ts_rmse = 10.24939, tr_c = 0.35616, val_c = 0.26227, ts_c = 0.25814
Epoch 657 time = 8.86, tr_rmse = 9.97326, val_rmse = 9.18643, ts_rmse = 9.23974, tr_c = 0.33288, val_c = 0.24726, ts_c = 0.24441
Epoch 658 time = 9.11, tr_rmse = 9.99359, val_rmse = 9.40147, ts_rmse = 9.47299, tr_c = 0.33465, val_c = 0.24999, ts_c = 0.24692
Epoch 659 time = 8.89, tr_rmse = 11.33285, val_rmse = 10.44514, ts_rmse = 10.32857, tr_c = 0.36556, val_c = 0.27383, ts_c = 0.26818
Epoch 660 time = 9.01, tr_rmse = 10.17247, val_rmse = 9.75069, ts_rmse = 9.81043, tr_c = 0.34104, val_c = 0.25479, ts_c = 0.25097
Epoch 661 time = 8.88, tr_rmse = 10.14547, val_rmse = 9.56890, ts_rmse = 9.59699, tr_c = 0.33581, val_c = 0.25044, ts_c = 0.24750
Epoch 662 time = 9.02, tr_rmse = 11.21248, val_rmse = 10.33856, ts_rmse = 10.25259, tr_c = 0.36527, val_c = 0.26795, ts_c = 0.26278
Epoch 663 time = 8.73, tr_rmse = 10.76279, val_rmse = 9.91318, ts_rmse = 9.82915, tr_c = 0.35268, val_c = 0.26277, ts_c = 0.25815
Epoch 664 time = 8.81, tr_rmse = 10.80954, val_rmse = 10.31416, ts_rmse = 10.39413, tr_c = 0.34901, val_c = 0.26329, ts_c = 0.26026
Epoch 665 time = 8.97, tr_rmse = 10.15278, val_rmse = 9.61361, ts_rmse = 9.67240, tr_c = 0.33601, val_c = 0.25051, ts_c = 0.24750
Epoch 666 time = 8.65, tr_rmse = 10.40875, val_rmse = 9.38066, ts_rmse = 9.32068, tr_c = 0.36932, val_c = 0.27896, ts_c = 0.27481
Epoch 667 time = 9.08, tr_rmse = 10.26058, val_rmse = 9.60206, ts_rmse = 9.47707, tr_c = 0.37242, val_c = 0.27613, ts_c = 0.26916
Epoch 668 time = 8.93, tr_rmse = 10.29008, val_rmse = 9.52632, ts_rmse = 9.60631, tr_c = 0.35198, val_c = 0.26050, ts_c = 0.25830
Epoch 669 time = 8.94, tr_rmse = 10.97051, val_rmse = 10.03803, ts_rmse = 10.01286, tr_c = 0.35937, val_c = 0.26689, ts_c = 0.26139
Epoch 670 time = 8.84, tr_rmse = 10.22319, val_rmse = 9.60865, ts_rmse = 9.68341, tr_c = 0.34065, val_c = 0.25352, ts_c = 0.25002
Epoch 671 time = 8.85, tr_rmse = 9.97551, val_rmse = 9.19892, ts_rmse = 9.25302, tr_c = 0.33497, val_c = 0.24971, ts_c = 0.24649
Epoch 672 time = 9.88, tr_rmse = 11.35845, val_rmse = 9.43841, ts_rmse = 9.51076, tr_c = 0.37093, val_c = 0.26884, ts_c = 0.26772
Epoch 673 time = 9.91, tr_rmse = 9.97187, val_rmse = 9.30898, ts_rmse = 9.36245, tr_c = 0.33810, val_c = 0.25111, ts_c = 0.24735
Epoch 674 time = 9.80, tr_rmse = 10.01479, val_rmse = 9.45651, ts_rmse = 9.49267, tr_c = 0.33302, val_c = 0.24823, ts_c = 0.24484
Epoch 675 time = 9.22, tr_rmse = 10.06929, val_rmse = 9.70848, ts_rmse = 9.76276, tr_c = 0.33688, val_c = 0.25279, ts_c = 0.24946
Epoch 676 time = 8.96, tr_rmse = 10.75566, val_rmse = 10.03612, ts_rmse = 10.05946, tr_c = 0.36283, val_c = 0.27209, ts_c = 0.26748
Epoch 677 time = 10.84, tr_rmse = 10.82458, val_rmse = 10.20095, ts_rmse = 10.15267, tr_c = 0.34388, val_c = 0.25613, ts_c = 0.25155
Epoch 678 time = 9.91, tr_rmse = 10.52713, val_rmse = 9.35514, ts_rmse = 9.43322, tr_c = 0.34206, val_c = 0.25281, ts_c = 0.25070
Epoch 679 time = 9.93, tr_rmse = 9.97220, val_rmse = 9.34884, ts_rmse = 9.42746, tr_c = 0.33628, val_c = 0.25170, ts_c = 0.24863
Epoch 680 time = 9.72, tr_rmse = 10.60701, val_rmse = 9.84598, ts_rmse = 9.84561, tr_c = 0.34229, val_c = 0.25474, ts_c = 0.25055
Epoch 681 time = 9.70, tr_rmse = 10.31652, val_rmse = 9.54120, ts_rmse = 9.50236, tr_c = 0.34368, val_c = 0.25264, ts_c = 0.24780
Epoch 682 time = 9.60, tr_rmse = 10.18568, val_rmse = 9.48085, ts_rmse = 9.50323, tr_c = 0.34025, val_c = 0.25029, ts_c = 0.24641
Epoch 683 time = 9.63, tr_rmse = 10.51845, val_rmse = 10.05414, ts_rmse = 10.00518, tr_c = 0.36263, val_c = 0.27014, ts_c = 0.26403
Epoch 684 time = 9.45, tr_rmse = 10.85813, val_rmse = 10.05507, ts_rmse = 9.93873, tr_c = 0.36799, val_c = 0.27127, ts_c = 0.26504
Epoch 685 time = 9.48, tr_rmse = 10.33319, val_rmse = 9.31879, ts_rmse = 9.42276, tr_c = 0.34676, val_c = 0.25698, ts_c = 0.25518
Epoch 686 time = 9.25, tr_rmse = 10.67490, val_rmse = 10.28641, ts_rmse = 10.36100, tr_c = 0.34734, val_c = 0.25926, ts_c = 0.25605
Epoch 687 time = 9.12, tr_rmse = 10.01224, val_rmse = 9.47536, ts_rmse = 9.52291, tr_c = 0.33570, val_c = 0.25012, ts_c = 0.24672
Epoch 688 time = 9.28, tr_rmse = 10.85141, val_rmse = 10.14378, ts_rmse = 10.03594, tr_c = 0.37271, val_c = 0.27083, ts_c = 0.26359
Epoch 689 time = 9.05, tr_rmse = 10.32716, val_rmse = 9.18625, ts_rmse = 9.25313, tr_c = 0.34393, val_c = 0.25228, ts_c = 0.25067
Epoch 690 time = 9.24, tr_rmse = 10.02979, val_rmse = 9.18719, ts_rmse = 9.24496, tr_c = 0.33852, val_c = 0.25201, ts_c = 0.24832
Epoch 691 time = 9.44, tr_rmse = 10.37609, val_rmse = 9.68697, ts_rmse = 9.77203, tr_c = 0.33653, val_c = 0.24952, ts_c = 0.24673
Epoch 692 time = 9.12, tr_rmse = 10.30415, val_rmse = 9.68349, ts_rmse = 9.64272, tr_c = 0.34494, val_c = 0.25910, ts_c = 0.25442
%% Cell type:code id: tags:
``` python
dst_net.eval()
out_r, out_c = dst_net(data_in_scaled[last_train:].to(device).float())
out_cc = F.softmax(out_c, dim=1)
out_cc = out_cc.detach().cpu().numpy()
from sklearn.metrics import confusion_matrix, matthews_corrcoef
from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error
np.set_printoptions(suppress=True, precision=3)
tp = 11
print(confusion_matrix(data_out_c[last_train:][:,tp].cpu().detach().numpy(), out_cc.argmax(axis=1).reshape(-1,12)[:,tp])/(confusion_matrix(data_out_c[last_train:][:,tp].cpu().detach().numpy(), out_cc.argmax(axis=1).reshape(-1,12)[:,tp]).sum(axis=1)[:, None]))
dst_levels = [-20,-50,-100]
truth = data_out[last_train:].cpu().detach().numpy().copy()
out = mm_scaler_out.inverse_transform(out_r.cpu().clone()).detach().numpy()
for i in range(12):
print(i, np.sqrt(mean_squared_error(truth[:,i], out[:,i])))
truth[np.where(truth >= dst_levels[0])] = 0
truth[np.where((truth < dst_levels[0]) & (truth >= dst_levels[1]))] = 1
truth[np.where((truth < dst_levels[1]) & (truth >= dst_levels[2]))] = 2
truth[np.where((truth < dst_levels[2]))] = 3
out[np.where(out >= dst_levels[0])] = 0
out[np.where((out < dst_levels[0]) & (out >= dst_levels[1]))] = 1
out[np.where((out < dst_levels[1]) & (out >= dst_levels[2]))] = 2
out[np.where((out < dst_levels[2]))] = 3
print(confusion_matrix(truth[:,11], out[:,11])/confusion_matrix(truth[:,11], out[:,11]).sum(axis=1)[:, None])
```
%%%% Output: stream
[[0.963 0.034 0.002 0.001]
[0.479 0.482 0.033 0.007]
[0.278 0.444 0.253 0.024]
[0.319 0.248 0.376 0.057]]
0 5.001647584723975
1 6.344738180233664
2 7.5513713736542645
3 8.30381312545836
4 9.049483066226928
5 9.566699248633029
6 10.072990258356135
7 10.593096784271411
8 11.03278087747998
9 11.487938816190386
10 11.835900790539245
11 12.123960219274442
[[0.961 0.036 0.002 0.001]
[0.497 0.455 0.045 0.004]
[0.27 0.399 0.306 0.025]
[0.255 0.255 0.454 0.035]]
%% Cell type:code id: tags:
``` python
```
......
......@@ -199,9 +199,9 @@ class DSTnet(nn.Module):
self.linear_o_3 = nn.Linear(self.nhidden_o, self.nhidden_o // 2)
self.linear_o_4_c = nn.Linear(self.nhidden_o // 2, self.after * 4)
self.linear_o_4_r = nn.Linear(4, 2)
# self.linear_o_4b_r = nn.Linear(2, 2)
self.linear_o_5_r = nn.Linear(2, 1)
self.linear_o_4_r = nn.Linear(4, 16)
self.linear_o_4b_r = nn.Linear(4, 4)
self.linear_o_5_r = nn.Linear(16, 1)
def init_hidden(self, batch_size):
......@@ -225,8 +225,8 @@ class DSTnet(nn.Module):
x = F.relu(self.linear_o_3(x))
x = F.dropout(x, 0.2, training=self.training)
x1 = self.linear_o_4_c(x)
x2 = self.linear_o_4_r(x1.view(-1, 4))
# x2 = self.linear_o_4b_r(x2)
x2 = F.relu(self.linear_o_4_r(x1.view(-1, 4)))
# x2 = F.relu(self.linear_o_4b_r(x2))
x2 = self.linear_o_5_r(x2)
x2 = x2.reshape(x0.size(0), self.after)
x1 = x1.reshape(x0.size(0) * self.after, 4)
......@@ -239,8 +239,8 @@ weights_c = torch.tensor([len(aa[aa==0])/len(aa[aa==0]), len(aa[aa==1])/len(aa[a
loss_f = nn.L1Loss()
loss_mse = nn.MSELoss(reduction='none')
#loss_fc= nn.CrossEntropyLoss()
loss_fc= nn.CrossEntropyLoss(weight = weights_c)
loss_fc= nn.CrossEntropyLoss()
# loss_fc= nn.CrossEntropyLoss(weight = weights_c)
nhidden_i = 2
nhidden_o = 96
......@@ -276,7 +276,8 @@ for epoch in range(num_epochs):
loss_r = loss_f(out_r, y_r)
loss_c = loss_fc(out_c, y_c)
loss = (loss_r * w).mean() + loss_c
# loss = (loss_r * w).mean() + 10 * loss_c
loss = (loss_r).mean() + loss_c
loss.backward()
optimizer.step()
......@@ -306,7 +307,7 @@ for epoch in range(num_epochs):
if (epoch % 10 == 0):
print('Epoch %d time = %.2f, tr_rmse = %0.5f, val_rmse = %0.5f, ts_rmse = %0.5f, tr_c = %.5f, val_c = %.5f, ts_c = %.5f' %
(epoch, epoch_time, loss_tr, loss_val, loss_ts, loss_c_tr, loss_c_val, loss_c_ts))
torch.save(dst_net.state_dict(), os.path.join('/home/mcristofo/DST/models','dst_regr_class_nsc.pth'))
......
......@@ -199,9 +199,9 @@ class DSTnet(nn.Module):
self.linear_o_3 = nn.Linear(self.nhidden_o, self.nhidden_o // 2)
self.linear_o_4_c = nn.Linear(self.nhidden_o // 2, self.after * 4)
self.linear_o_4_r = nn.Linear(4, 8)
self.linear_o_4b_r = nn.Linear(8, 4)
self.linear_o_5_r = nn.Linear(4, 1)
self.linear_o_4_r = nn.Linear(4, 16)
self.linear_o_4b_r = nn.Linear(4, 4)
self.linear_o_5_r = nn.Linear(16, 1)
def init_hidden(self, batch_size):
......@@ -226,7 +226,7 @@ class DSTnet(nn.Module):
x = F.dropout(x, 0.2, training=self.training)
x1 = self.linear_o_4_c(x)
x2 = F.relu(self.linear_o_4_r(x1.view(-1, 4)))
x2 = F.relu(self.linear_o_4b_r(x2))
# x2 = F.relu(self.linear_o_4b_r(x2))
x2 = self.linear_o_5_r(x2)
x2 = x2.reshape(x0.size(0), self.after)
x1 = x1.reshape(x0.size(0) * self.after, 4)
......@@ -239,8 +239,8 @@ weights_c = torch.tensor([len(aa[aa==0])/len(aa[aa==0]), len(aa[aa==1])/len(aa[a
loss_f = nn.L1Loss()
loss_mse = nn.MSELoss(reduction='none')
#loss_fc= nn.CrossEntropyLoss()
loss_fc= nn.CrossEntropyLoss(weight = weights_c)
loss_fc= nn.CrossEntropyLoss()
# loss_fc= nn.CrossEntropyLoss(weight = weights_c)
nhidden_i = 2
nhidden_o = 96
......@@ -276,7 +276,8 @@ for epoch in range(num_epochs):
loss_r = loss_f(out_r, y_r)
loss_c = loss_fc(out_c, y_c)
loss = (loss_r * w).mean() + loss_c
# loss = (loss_r * w).mean() + 10 * loss_c
loss = (loss_r).mean() + loss_c
loss.backward()
optimizer.step()
......@@ -306,7 +307,7 @@ for epoch in range(num_epochs):
if (epoch % 10 == 0):
print('Epoch %d time = %.2f, tr_rmse = %0.5f, val_rmse = %0.5f, ts_rmse = %0.5f, tr_c = %.5f, val_c = %.5f, ts_c = %.5f' %
(epoch, epoch_time, loss_tr, loss_val, loss_ts, loss_c_tr, loss_c_val, loss_c_ts))
torch.save(dst_net.state_dict(), os.path.join('/home/mcristofo/DST/models','dst_regr_class_nsc.pth'))
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment