amorgun

import pandas as pd
import numpy as np
from sklearn.feature_extraction import DictVectorizer
 
def encode_onehot(df, cols):
    """
    One-hot encoding is applied to columns specified in a pandas DataFrame.
    
    Modified from: https://gist.github.com/kljensen/5452382
    

amorgun

virtualenv --no-site-packages --distribute -p `which python3` <env_name>

or simply

pyvenv <env_name>

amorgun

sudo kill -9 $(ps aux | grep <name> | awk '{print $2}')

amorgun

import tkinter
import time

master = tkinter.Tk()

canvas = tkinter.Canvas(master, width=1000, height=1000)
canvas.pack()

def animation():
    def draw_point(x, y, c='black'):

amorgun

import numpy as np
import scipy as sp
import scipy.sparse
import tempfile


def hstack(parts):
   with tempfile.TemporaryFile() as data_file, tempfile.TemporaryFile() as indices_file:
       data = np.memmap(data_file,
                        dtype=parts[0].dtype,

amorgun

from scipy.sparse import csr_matrix
import numpy as np

def make_sparse(clf):
    """
    Make sklearn.svm.SVC trained on dense data work on sparse features without fitting if again.
    """
    clf._sparse = True
    clf.support_vectors_ = csr_matrix(clf.support_vectors_)

amorgun

from sqlalchemy.dialects import postgresql

def bulk_upsert(session: Session, 
                items: Sequence[Mapping[str, Any]]):
  session.execute(
    postgresql.insert(MyModel.__table__)
    .values(items)
    .on_conflict_do_update(
      index_elements=[MyModel.id],
      set_={MyModel.my_field.name: 'new_value'},

amorgun

# After Ubuntu 16.04, Systemd becomes the default. 
# It is simpler than https://gist.github.com/Doowon/38910829898a6624ce4ed554f082c4dd

[Unit]
Description=Jupyter Notebook

[Service]
Type=simple
PIDFile=/run/jupyter.pid
ExecStart=/home/phil/Enthought/Canopy_64bit/User/bin/jupyter-notebook --config=/home/phil/.jupyter/jupyter_notebook_config.py

amorgun

# register file moved without svn mv
set -eu
shopt -s expand_aliases
. ~/.bash_aliases
old=$1
new=$2
tmp=$(mktemp svnmove.XXXXXX)
svn rm --keep-local "$new"
mv "$new" "$tmp"
svn revert "$old"

amorgun

def print_stats(y_true, y_pred):
    print(f'Total accuracy: {accuracy_score(y_true, y_pred)}')
    print()
    cm = pd.DataFrame(confusion_matrix(y_true, y_pred),
                      columns=pd.MultiIndex.from_arrays([['not_fit', 'fit']], names=['My fit']),
                      index=['not_fit', 'fit'],
                 )
    cm.index.name = 'Toloka fit'
    print('Confusion matrix')
    print(cm)