satomacoto

import base64
import cv2
from PIL import Image

cap = cv2.VideoCapture(0)
_, img = cap.read()
image_string = cv2.imencode(".png", img)[1].tostring()

output = io.BytesIO()
Image.fromarray(img).save(output, format='PNG')

satomacoto

docker images | awk '/<none/{print $3}' | xargs docker rmi
docker ps -f "status=exited" -q | xargs docker rm

satomacoto

curl -fsSL https://gist.githubusercontent.com/satomacoto/b27dc3f871938d2614819bba3c589b45/raw/setup1.sh | bash
curl -fsSL https://gist.githubusercontent.com/satomacoto/b27dc3f871938d2614819bba3c589b45/raw/setup2.sh | bash

satomacoto

pyenv install miniconda3-latest
pyenv local miniconda3-latest
conda create -n myenv anaconda
source (conda info --root)/etc/fish/conf.d/conda.fish
conda activate myenv

satomacoto

# create an account, create an app
# @see https://apps.twitter.com/

# retrieve the access tokens
# @see https://dev.twitter.com/oauth/reference/post/oauth2/token

# create the file ~/twitter_api
nano ~/twitter_api
Authorization: OAuth oauth_consumer_key="XXXXXX", oauth_nonce="11111111", oauth_signature="XXXXXX", oauth_signature_method="HMAC-SHA1", oauth_timestamp="1450728725", oauth_token="99999-XXXXXX", oauth_version="1.0"

satomacoto

機械学習ライブラリを利用するとき、モデルやパラメータの良し悪しを判断するための評価基準。

• 3.3. Model evaluation_ quantifying the quality of predictions — scikit-learn 0.17.1 documentation
• 3.5. Validation curves_ plotting scores to evaluate models — scikit-learn 0.17.1 documentation
• Machine Learning with Scikit Learn (Part II) - 人工知能に関する断創録

4.1 Cross validation

4.2 Model Complexity and GridSearchCV

satomacoto

{"nodes":[{"name":"Myriel","group":1},{"name":"Napoleon","group":1},{"name":"Mlle.Baptistine","group":1},{"name":"Mme.Magloire","group":1},{"name":"CountessdeLo","group":1},{"name":"Geborand","group":1},{"name":"Champtercier","group":1},{"name":"Cravatte","group":1},{"name":"Count","group":1},{"name":"OldMan","group":1},{"name":"Labarre","group":2},{"name":"Valjean","group":2},{"name":"Marguerite","group":3},{"name":"Mme.deR","group":2},{"name":"Isabeau","group":2},{"name":"Gervais","group":2},{"name":"Tholomyes","group":3},{"name":"Listolier","group":3},{"name":"Fameuil","group":3},{"name":"Blacheville","group":3},{"name":"Favourite","group":3},{"name":"Dahlia","group":3},{"name":"Zephine","group":3},{"name":"Fantine","group":3},{"name":"Mme.Thenardier","group":4},{"name":"Thenardier","group":4},{"name":"Cosette","group":5},{"name":"Javert","group":4},{"name":"Fauchelevent","group":0},{"name":"Bamatabois","group":2},{"name":"Perpetue","group":3},{"name":"Simplice","group":2},{"name":"Scaufflaire","group":2},

satomacoto

# -*- coding: utf-8 -*-

from sklearn.base import BaseEstimator, RegressorMixin, TransformerMixin
from sklearn.cross_validation import cross_val_predict


class ThroughTransformer(BaseEstimator, TransformerMixin):
    def fit(self, X, y=None, **fit_param):
        return self

satomacoto

Install pyenv

$ brew update
$ brew install pyenv

Check available versions and install

satomacoto

class Pinv2Regressor:
    def __init__(self, estimators, cv=5):
        self.estimators = estimators
        self.cv = cv

    def fit(self, X, y, verbose=False):
        preds = {}
        for k, estimator in self.estimators.items():
            if verbose:
                print(k, estimator)