DeepMarket/main.py at main · LeonardoBerti00/DeepMarket · GitHub

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
import random
import numpy as np
import wandb
from run import run_wandb, run, sweep_init
import torch
import constants as cst
import configuration
import warnings
from preprocessing.LOBSTERDataBuilder import LOBSTERDataBuilder
import evaluation.quantitative_eval.predictive_lstm as predictive_lstm
import evaluation.visualizations.comparison_distribution_order_type as comparison_distribution_order_type
import evaluation.visualizations.comparison_distribution_market_spread as comparison_distribution_market_spread
import evaluation.visualizations.PCA_plots as PCA_plots
import evaluation.visualizations.comparison_midprice as comparison_midprice
import evaluation.visualizations.comparison_volume_distribution as comparison_volume_distribution
import evaluation.visualizations.comparison_core_coef_lags as comparison_core_coef_lags
import evaluation.visualizations.comparison_correlation_coefficient as comparison_correlation_coefficient
import evaluation.visualizations.comparison_log_return_frequency as comparison_log_return_frequency
import evaluation.visualizations.comparison_distribution_log_interarrival_times as comparison_distribution_log_interarrival_times

def set_repoducibility():
    torch.manual_seed(cst.SEED)
    np.random.seed(cst.SEED)
    random.seed(cst.SEED)

def set_torch():
    torch.set_default_dtype(torch.float32)
    torch.backends.cuda.matmul.allow_tf32 = True
    torch.backends.cudnn.allow_tf32 = True
    torch.autograd.set_detect_anomaly(False)
    torch.set_float32_matmul_precision('high')
    # this is done for tensorflow
    import os
    os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

def plot_graphs(real_data_path=None, TRADES_data_path=None, iabs_data_path=None, cgan_data_path=None):
    warnings.filterwarnings("ignore")
    comparison_distribution_order_type.main(real_data_path, TRADES_data_path, iabs_data_path, cgan_data_path)
    comparison_distribution_market_spread.main(real_data_path, TRADES_data_path, iabs_data_path, cgan_data_path)
    PCA_plots.main(real_data_path, TRADES_data_path)
    comparison_midprice.main(real_data_path, TRADES_data_path)
    comparison_volume_distribution.main(TRADES_data_path)
    # these last plots are slow, they will take a couple of minutes to run
    comparison_core_coef_lags.main(real_data_path, TRADES_data_path, iabs_data_path, cgan_data_path)
    comparison_correlation_coefficient.main(real_data_path, TRADES_data_path, iabs_data_path, cgan_data_path)
    comparison_log_return_frequency.main(real_data_path, TRADES_data_path, cgan_data_path)
    comparison_distribution_log_interarrival_times.main(real_data_path, TRADES_data_path, iabs_data_path, cgan_data_path)

if __name__ == "__main__":
    set_torch()
    set_repoducibility()

    config = configuration.Configuration()
    if (cst.DEVICE == "cpu"):
        accelerator = "cpu"
    else:
        accelerator = "gpu"

    if (not config.IS_DATA_PREPROCESSED):
        for i in range(len(config.CHOSEN_STOCK)):
            # prepare the datasets, this will save train.npy, val.npy and test.npy in the data directory
            data_builder = LOBSTERDataBuilder(
                stock_name=config.CHOSEN_STOCK[i].name,
                data_dir=cst.DATA_DIR,
                date_trading_days=cst.DATE_TRADING_DAYS,
                split_rates=config.SPLIT_RATES,
                chosen_model=config.CHOSEN_MODEL
            )
            data_builder.prepare_save_datasets()

    if config.IS_WANDB:
        if config.IS_SWEEP:
            sweep_config = sweep_init(config)
            sweep_id = wandb.sweep(sweep_config, project=cst.PROJECT_NAME, entity="")
            wandb.agent(sweep_id, run_wandb(config, accelerator), count=sweep_config["run_cap"])
        else:
            start_wandb = run_wandb(config, accelerator)
            start_wandb()

    # training without using wandb
    elif config.IS_TRAINING:
        run(config, accelerator)

    elif config.IS_EVALUATION:
        plot_graphs(config.REAL_DATA_PATH, config.TRADES_DATA_PATH, config.IABS_DATA_PATH, config.CGAN_DATA_PATH)
        predictive_lstm.main(config.REAL_DATA_PATH, config.TRADES_DATA_PATH)