psinger

hardstatus alwayslastline
hardstatus string '%{= kG}[%{G}%H%? %1`%?%{g}][%= %{= kw}%-w%{+b yk} %n*%t%?(%u)%? %{-}%+w %=%{g}][%{B}%m/%d %{W}%C%A%{g}]'

# huge scrollback buffer
defscrollback 5000

# no welcome message
startup_message off

# 256 colors

psinger

""" Calibrated Classifier Model: To calibrate predictions using Platt's scaling, Isotonic Regression or Splines
"""

import copy
import datatable as dt
from h2oaicore.mojo import MojoWriter, MojoFrame
from h2oaicore.systemutils import config
from h2oaicore.models import CustomModel, LightGBMModel
from sklearn.preprocessing import LabelEncoder
import numpy as np

psinger

""" Calibrated Classifier Model: To calibrate predictions using Platt's scaling, Isotonic Regression or Splines
"""

import copy
import datatable as dt
from h2oaicore.mojo import MojoWriter, MojoFrame
from h2oaicore.systemutils import config
from h2oaicore.models import CustomModel, LightGBMModel
from sklearn.preprocessing import LabelEncoder
import numpy as np

psinger

                Name  # plays  Actual  Expected  Ratio    Dif
     Kerryon Johnson      118   5.432     3.755  1.446  1.677
       Kenneth Dixon       60   5.550     4.201  1.321  1.349
      Saquon Barkley      261   5.008     3.738  1.340  1.270
          Nick Chubb      190   5.184     3.931  1.319  1.253
       Derrick Henry      200   4.825     3.695  1.306  1.130
      Jordan Wilkins       60   5.600     4.558  1.229  1.042
         Aaron Jones      132   5.485     4.465  1.228  1.020
         Zach Zenner       55   4.818     3.845  1.253  0.973
     Phillip Lindsay      189   5.466     4.510  1.212  0.955

psinger

                Name  # plays  Actual  Expected  Ratio    Dif
     Kerryon Johnson      118   5.432     3.755  1.446  1.677
       Kenneth Dixon       60   5.550     4.201  1.321  1.349
      Saquon Barkley      261   5.008     3.738  1.340  1.270
          Nick Chubb      190   5.184     3.931  1.319  1.253
       Derrick Henry      200   4.825     3.695  1.306  1.130
      Jordan Wilkins       60   5.600     4.558  1.229  1.042
         Aaron Jones      132   5.485     4.465  1.228  1.020
         Zach Zenner       55   4.818     3.845  1.253  0.973
     Phillip Lindsay      189   5.466     4.510  1.212  0.955

psinger

from joblib import Parallel, delayed
import multiprocessing
import pandas as pd
import time
def applyParallel(dfGrouped, func):
    retLst = Parallel(n_jobs=multiprocessing.cpu_count())(delayed(func)(group) for name, group in dfGrouped)
    return pd.concat(retLst)
def myfunc(df)
    return df
  

psinger

#https://github.com/qubvel/ttach
import ttach as tta

transforms = tta.Compose(
  [
    tta.HorizontalFlip(),
    tta.VerticalFlip(),
    #                 tta.Rotate90(angles=[0, 180]),
    #                 tta.Scale(scales=[1, 2, 4]),
    #                 tta.Multiply(factors=[0.9, 1, 1.1]),        

psinger

from sklearn.base import BaseEstimator, TransformerMixin
from pandas.api.types import CategoricalDtype
import pandas as pd


class DummyEncoder(BaseEstimator, TransformerMixin):
    def __init__(self, min_frequency=1, dummy_na=True):
        self.min_frequency = min_frequency
        self.dummy_na = dummy_na
        self.categories = dict()

psinger

import numpy as np
x = np.array([525., 300., 450., 300., 400., 500., 550., 125., 300., 400., 500., 550.])
y = np.array([250., 225., 275., 350., 325., 375., 450., 400., 500., 550., 600., 525.])
data = np.array([x, y])
model = analyze(data)

psinger

from pymc import Normal, Uniform, MvNormal, Exponential
from numpy.linalg import inv, det
from numpy import log, pi, dot
import numpy as np
from scipy.special import gammaln
 
def _model(data, robust=False):
    # priors might be adapted here to be less flat
    mu = Normal('mu', 0, 0.000001, size=2)
    sigma = Uniform('sigma', 0, 1000, size=2)