forked from Seanny123/da-rnn
-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
218 lines (164 loc) · 8.41 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
import typing
from typing import Tuple
import json
import os
import torch
from torch import nn
from torch import optim
from sklearn.preprocessing import StandardScaler
from sklearn.externals import joblib
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import utils
from modules import Encoder, Decoder
from custom_types import DaRnnNet, TrainData, TrainConfig
from utils import numpy_to_tvar
from constants import device
logger = utils.setup_log()
logger.info(f"Using computation device: {device}")
def preprocess_data(dat, col_names) -> Tuple[TrainData, StandardScaler]:
scale = StandardScaler().fit(dat)
proc_dat = scale.transform(dat)
mask = np.ones(proc_dat.shape[1], dtype=bool)
dat_cols = list(dat.columns)
for col_name in col_names:
mask[dat_cols.index(col_name)] = False
feats = proc_dat[:, mask]
targs = proc_dat[:, ~mask]
return TrainData(feats, targs), scale
def da_rnn(train_data: TrainData, n_targs: int, encoder_hidden_size=64, decoder_hidden_size=64,
T=10, learning_rate=0.01, batch_size=128):
train_cfg = TrainConfig(T, int(train_data.feats.shape[0] * 0.7), batch_size, nn.MSELoss())
logger.info(f"Training size: {train_cfg.train_size:d}.")
enc_kwargs = {"input_size": train_data.feats.shape[1], "hidden_size": encoder_hidden_size, "T": T}
encoder = Encoder(**enc_kwargs).to(device)
with open(os.path.join("data", "enc_kwargs.json"), "w") as fi:
json.dump(enc_kwargs, fi, indent=4)
dec_kwargs = {"encoder_hidden_size": encoder_hidden_size,
"decoder_hidden_size": decoder_hidden_size, "T": T, "out_feats": n_targs}
decoder = Decoder(**dec_kwargs).to(device)
with open(os.path.join("data", "dec_kwargs.json"), "w") as fi:
json.dump(dec_kwargs, fi, indent=4)
encoder_optimizer = optim.Adam(
params=[p for p in encoder.parameters() if p.requires_grad],
lr=learning_rate)
decoder_optimizer = optim.Adam(
params=[p for p in decoder.parameters() if p.requires_grad],
lr=learning_rate)
da_rnn_net = DaRnnNet(encoder, decoder, encoder_optimizer, decoder_optimizer)
return train_cfg, da_rnn_net
def train(net: DaRnnNet, train_data: TrainData, t_cfg: TrainConfig, n_epochs=10, save_plots=False):
iter_per_epoch = int(np.ceil(t_cfg.train_size * 1. / t_cfg.batch_size))
iter_losses = np.zeros(n_epochs * iter_per_epoch)
epoch_losses = np.zeros(n_epochs)
logger.info(f"Iterations per epoch: {t_cfg.train_size * 1. / t_cfg.batch_size:3.3f} ~ {iter_per_epoch:d}.")
n_iter = 0
for e_i in range(n_epochs):
perm_idx = np.random.permutation(t_cfg.train_size - t_cfg.T)
for t_i in range(0, t_cfg.train_size, t_cfg.batch_size):
batch_idx = perm_idx[t_i:(t_i + t_cfg.batch_size)]
feats, y_history, y_target = prep_train_data(batch_idx, t_cfg, train_data)
loss = train_iteration(net, t_cfg.loss_func, feats, y_history, y_target)
iter_losses[e_i * iter_per_epoch + t_i // t_cfg.batch_size] = loss
# if (j / t_cfg.batch_size) % 50 == 0:
# self.logger.info("Epoch %d, Batch %d: loss = %3.3f.", i, j / t_cfg.batch_size, loss)
n_iter += 1
adjust_learning_rate(net, n_iter)
epoch_losses[e_i] = np.mean(iter_losses[range(e_i * iter_per_epoch, (e_i + 1) * iter_per_epoch)])
if e_i % 10 == 0:
y_test_pred = predict(net, train_data,
t_cfg.train_size, t_cfg.batch_size, t_cfg.T,
on_train=False)
# TODO: make this MSE and make it work for multiple inputs
val_loss = y_test_pred - train_data.targs[t_cfg.train_size:]
logger.info(f"Epoch {e_i:d}, train loss: {epoch_losses[e_i]:3.3f}, val loss: {np.mean(np.abs(val_loss))}.")
y_train_pred = predict(net, train_data,
t_cfg.train_size, t_cfg.batch_size, t_cfg.T,
on_train=True)
plt.figure()
plt.plot(range(1, 1 + len(train_data.targs)), train_data.targs,
label="True")
plt.plot(range(t_cfg.T, len(y_train_pred) + t_cfg.T), y_train_pred,
label='Predicted - Train')
plt.plot(range(t_cfg.T + len(y_train_pred), len(train_data.targs) + 1), y_test_pred,
label='Predicted - Test')
plt.legend(loc='upper left')
utils.save_or_show_plot(f"pred_{e_i}.png", save_plots)
return iter_losses, epoch_losses
def prep_train_data(batch_idx: np.ndarray, t_cfg: TrainConfig, train_data: TrainData):
feats = np.zeros((len(batch_idx), t_cfg.T - 1, train_data.feats.shape[1]))
y_history = np.zeros((len(batch_idx), t_cfg.T - 1, train_data.targs.shape[1]))
y_target = train_data.targs[batch_idx + t_cfg.T]
for b_i, b_idx in enumerate(batch_idx):
b_slc = slice(b_idx, b_idx + t_cfg.T - 1)
feats[b_i, :, :] = train_data.feats[b_slc, :]
y_history[b_i, :] = train_data.targs[b_slc]
return feats, y_history, y_target
def adjust_learning_rate(net: DaRnnNet, n_iter: int):
# TODO: Where did this Learning Rate adjustment schedule come from?
# Should be modified to use Cosine Annealing with warm restarts https://www.jeremyjordan.me/nn-learning-rate/
if n_iter % 10000 == 0 and n_iter > 0:
for enc_params, dec_params in zip(net.enc_opt.param_groups, net.dec_opt.param_groups):
enc_params['lr'] = enc_params['lr'] * 0.9
dec_params['lr'] = dec_params['lr'] * 0.9
def train_iteration(t_net: DaRnnNet, loss_func: typing.Callable, X, y_history, y_target):
t_net.enc_opt.zero_grad()
t_net.dec_opt.zero_grad()
input_weighted, input_encoded = t_net.encoder(numpy_to_tvar(X))
y_pred = t_net.decoder(input_encoded, numpy_to_tvar(y_history))
y_true = numpy_to_tvar(y_target)
loss = loss_func(y_pred, y_true)
loss.backward()
t_net.enc_opt.step()
t_net.dec_opt.step()
return loss.item()
def predict(t_net: DaRnnNet, t_dat: TrainData, train_size: int, batch_size: int, T: int, on_train=False):
out_size = t_dat.targs.shape[1]
if on_train:
y_pred = np.zeros((train_size - T + 1, out_size))
else:
y_pred = np.zeros((t_dat.feats.shape[0] - train_size, out_size))
for y_i in range(0, len(y_pred), batch_size):
y_slc = slice(y_i, y_i + batch_size)
batch_idx = range(len(y_pred))[y_slc]
b_len = len(batch_idx)
X = np.zeros((b_len, T - 1, t_dat.feats.shape[1]))
y_history = np.zeros((b_len, T - 1, t_dat.targs.shape[1]))
for b_i, b_idx in enumerate(batch_idx):
if on_train:
idx = range(b_idx, b_idx + T - 1)
else:
idx = range(b_idx + train_size - T, b_idx + train_size - 1)
X[b_i, :, :] = t_dat.feats[idx, :]
y_history[b_i, :] = t_dat.targs[idx]
y_history = numpy_to_tvar(y_history)
_, input_encoded = t_net.encoder(numpy_to_tvar(X))
y_pred[y_slc] = t_net.decoder(input_encoded, y_history).cpu().data.numpy()
return y_pred
save_plots = True
debug = False
raw_data = pd.read_csv(os.path.join("data", "nasdaq100_padding.csv"), nrows=100 if debug else None)
logger.info(f"Shape of data: {raw_data.shape}.\nMissing in data: {raw_data.isnull().sum().sum()}.")
targ_cols = ("NDX",)
data, scaler = preprocess_data(raw_data, targ_cols)
da_rnn_kwargs = {"batch_size": 128, "T": 10}
config, model = da_rnn(data, n_targs=len(targ_cols), learning_rate=.001, **da_rnn_kwargs)
iter_loss, epoch_loss = train(model, data, config, n_epochs=10, save_plots=save_plots)
final_y_pred = predict(model, data, config.train_size, config.batch_size, config.T)
plt.figure()
plt.semilogy(range(len(iter_loss)), iter_loss)
utils.save_or_show_plot("iter_loss.png", save_plots)
plt.figure()
plt.semilogy(range(len(epoch_loss)), epoch_loss)
utils.save_or_show_plot("epoch_loss.png", save_plots)
plt.figure()
plt.plot(final_y_pred, label='Predicted')
plt.plot(data.targs[config.train_size:], label="True")
plt.legend(loc='upper left')
utils.save_or_show_plot("final_predicted.png", save_plots)
with open(os.path.join("data", "da_rnn_kwargs.json"), "w") as fi:
json.dump(da_rnn_kwargs, fi, indent=4)
joblib.dump(scaler, os.path.join("data", "scaler.pkl"))
torch.save(model.encoder.state_dict(), os.path.join("data", "encoder.torch"))
torch.save(model.decoder.state_dict(), os.path.join("data", "decoder.torch"))