casperkaae

def update_dynamic_datasets(outputfolder=None, recompute_out_of_sample_static_y_hat=True):
    ###### COMBINED PREPREOCESSING ###########
    logger.info('Preprocessing - making out of sample predictions')
    target_col = 'intraday_pct_change_next'
    #df = load_newest_dataset_from_fileserver(STATIC_EARNINGS_DATA_NAME)
    # ds = load_newest_dataset_from_fileserver(MARKET_BAR_DATASET_NAME)
    #df = get_static_earnings_input_data(df, ds, overwrite_overnight_pct_change=True)

    #if settings.STATIC_DATASET_SETTINGS['use_shifted_static_data'] and recompute_out_of_sample_static_y_hat:
    #    df_shift = load_newest_dataset_from_fileserver(STATIC_EARNINGS_DATA_SHIFTED_NAME)

casperkaae

# Packages we're using
import numpy as np
import copy
import soundfile as sf

def overlap(X, window_size, window_step):
    """
    Create an overlapped version of X
    Parameters

casperkaae

# Packages we're using
import numpy as np
import matplotlib.pyplot as plt
import copy
from scipy.io import wavfile
import soundfile as sf

from utils.snr import residual_snrdb

casperkaae

import numpy as np

import torch
from torch.autograd import Variable

def dynamic_avg_pooling_test():
    
    from functions.dynamic_avg_pooling import dynamic_avg_pooling
    batch_size = 7 #crashes if below 8
    nc = 32

casperkaae

H=heads, T=Tail

HHH = 1
HHT = 2
HTH = 3
HTT = 4
THH = 5
THT = flip again
TTH = flip again
TTT = flip again

casperkaae

def mirror_padding(images, filter_size):
    """
    Mirror padding is used to apply a 2D convolution avoiding the border
    effects that one normally gets with zero padding.

    We assume that the filter has an odd size.

    To obtain a filtered tensor with the same output size, substitute
    a ``conv2d(images, filters, mode="half")`` with
    ``conv2d(mirror_padding(images, filters.shape), filters, mode="valid")``.

casperkaae

class LinearLayer():
    def __init__(self, num_inputs, num_units, scale=0.01):
        self.num_units = num_units
        self.num_inputs = num_inputs
        self.W = np.random.random((num_inputs, num_units)) * scale
        self.b = np.zeros(num_units)

    def __str__(self): 
        return "LinearLayer(%i, %i)" % (self.num_inputs, self.num_units)