kingjr

import numpy as np
from sklearn.preprocessing import RobustScaler
from sklearn.base import TransformerMixin

class TanhScaler(TransformerMixin):
    def __init__(self,
                 factor=2., 
                 decim=1,
                 n_max=None,
                 shuffle=False,

kingjr

class TimeGenScorer:
    def __init__(self, n_times, train=None, cuda=False):
        self.test_times = range(n_times)
        if train is None:
            train = slice(None, None, 1)
        self.train_times = self.test_times[train]
        self.cuda = cuda

    def __call__(self, Y_true, Y_pred):

kingjr

def stats(X, n_jobs=-1, seed=0, **kwargs):
    from mne.stats import spatio_temporal_cluster_1samp_test
    X = np.asarray(X)
    if X.ndim==2:
        X = X[:, :, None]
    _, clusters, c_pv, _ = spatio_temporal_cluster_1samp_test(
        X, n_jobs=n_jobs, seed=seed, out_type='mask', **kwargs
    )
    p_vals = [(mask*p + (mask==0)) for mask, p in zip(clusters, c_pv)
              if p < .05]

kingjr

def scale(X):
    m = X.mean(0, keepdims=True)
    s = X.std(0, keepdims=True)
    s[s == 0] = 1  # avoid nan
    X -= m
    X /= s
    return X

kingjr

import mne
import numpy as np
import matplotlib.pyplot as plt
import scipy
import torch
from itertools import product
import pandas as pd
import seaborn as sns

def cart_to_pol(x, y):

kingjr

kingjr

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",

kingjr

from sklearn.linear_model import RidgeCV
import numpy as np
from scipy import linalg


class BlockRidgeCV(RidgeCV):
    def __init__(self, alphas, blocks, fit_intercept=False, **kwargs):
        super().__init__(**kwargs)
        self.fit_intercept = False
        self.alphas = alphas

kingjr

import matplotlib.pyplot as plt
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.cross_decomposition import PLSRegression
from sklearn.model_selection import cross_val_score
from scipy.stats import pearsonr

ols = LinearRegression()
pls = PLSRegression()

kingjr

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np


n_layers = 4
n_steps = 4


fig, axes = plt.subplots(1, 3, sharex=True, sharey=True, figsize=[10, 3])