今日の作業記録 python error（言語処理100本ノック:79）未解決

言語処理100本ノック 2015

79. 適合率-再現率グラフの描画

http://www.cl.ecei.tohoku.ac.jp/nlp100/
「アーティスト情報（artist.json.gz）をデータベースに登録せよ．さらに，次のフィールドでインデックスを作成せよ: name, aliases.name, tags.value, rating.value．」
素人の言語処理100本ノック:79
https://qiita.com/segavvy/items/8f93187ec89f4831d863

# ./p79.py
Traceback (most recent call last):
  File "./p79.py", line 105, in <module>
    plt.plot(thresholds, accuracys, color='green', linestyle='--', label='正解率')
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/pyplot.py", line 3352, in plot
    ax = gca()
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/pyplot.py", line 969, in gca
    return gcf().gca(**kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/pyplot.py", line 586, in gcf
    return figure()
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/pyplot.py", line 533, in figure
    **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backend_bases.py", line 161, in new_figure_manager
    return cls.new_figure_manager_given_figure(num, fig)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backend_bases.py", line 167, in new_figure_manager_given_figure
    canvas = cls.FigureCanvas(figure)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_qt5agg.py", line 24, in __init__
    super(FigureCanvasQTAgg, self).__init__(figure=figure)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_qt5.py", line 234, in __init__
    _create_qApp()
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_qt5.py", line 125, in _create_qApp
    raise RuntimeError('Invalid DISPLAY variable')
RuntimeError: Invalid DISPLAY variable

ソースは下記（コマンドとして実行したく１行目追記）

#!/usr/bin/env python
# coding: utf-8

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.font_manager import FontProperties

fname_result = 'result.txt'
fname_work = 'work.txt'


def score(fname):
    '''結果ファイルからスコア算出
    結果ファイルを読み込んで、正解率、適合率、再現率、F1スコアを返す

    戻り値：
    正解率,適合率,再現率,F1スコア
    '''
    # 結果を読み込んで集計
    TP = 0      # True-Positive     予想が+1、正解も+1
    FP = 0      # False-Positive    予想が+1、正解は-1
    FN = 0      # False-Negative    予想が-1、正解は+1
    TN = 0      # True-Negative     予想が-1、正解も-1

    with open(fname) as data_file:
        for line in data_file:
            cols = line.split('\t')

            if len(cols) < 3:
                continue

            if cols[0] == '+1':         # 正解
                if cols[1] == '+1':     # 予想
                    TP += 1
                else:
                    FN += 1
            else:
                if cols[1] == '+1':
                    FP += 1
                else:
                    TN += 1

    # 算出
    accuracy = (TP + TN) / (TP + FP + FN + TN)      # 正解率
    precision = TP / (TP + FP)      # 適合率
    recall = TP / (TP + FN)     # 再現率
    f1 = (2 * recall * precision) / (recall + precision)    # F1スコア

    return accuracy, precision, recall, f1


# 結果読み込み、予測確率は元の値（仮説関数hypothesis()の値）に戻す
results = []
with open(fname_result) as data_file:
    for line in data_file:

        cols = line.split('\t')
        if len(cols) < 3:
            continue

        # 正解ラベル
        label = cols[0]

        # 識別関数predict()の値
        if cols[1] == '-1':
            predict = 1.0 - float(cols[2])      # 確率を戻す
        else:
            predict = float(cols[2])

        results.append((label, predict))

# 閾値を変えながらスコア算出、グラフ描画用の配列へセット
thresholds = []
accuracys = []
precisions = []
recalls = []
f1s = []
for threshold in np.arange(0.02, 1.0, 0.02):

    # score()を使うため、一時ファイルに結果保存
    with open(fname_work, 'w') as file_out:
        for label, predict in results:
            if predict > threshold:
                file_out.write('{}\t{}\t{}\n'.format(label, '+1', predict))
            else:
                file_out.write('{}\t{}\t{}\n'.format(label, '-1', 1 - predict))

    # スコア算出
    accuracy, precision, recall, f1 = score(fname_work)

    # 結果追加
    thresholds.append(threshold)
    accuracys.append(accuracy)
    precisions.append(precision)
    recalls.append(recall)
    f1s.append(f1)


# グラフで使うフォント情報(デフォルトのままでは日本語が表示できない)
fp = FontProperties(
    fname='/Library/Fonts/Times New Roman Bold Italic.ttf'
)

# 折線グラフの値の設定
plt.plot(thresholds, accuracys, color='green', linestyle='--', label='正解率')
plt.plot(thresholds, precisions, color='red', linewidth=3, label='適合率')
plt.plot(thresholds, recalls, color='blue', linewidth=3, label='再現率')
plt.plot(thresholds, f1s, color='magenta', linestyle='--', label='F1スコア')

# 軸の値の範囲の調整
plt.xlim(
    xmin=0, xmax=1.0
)
plt.ylim(
    ymin=0, ymax=1.0
)

# グラフのタイトル、ラベル指定
plt.title(
    '79. 適合率-再現率グラフの描画',    # タイトル
    fontproperties=fp   # 使うフォント情報
)
plt.xlabel(
    'ロジスティック回帰モデルの分類の閾値',       # x軸ラベル
    fontproperties=fp   # 使うフォント情報
)
plt.ylabel(
    '精度',         # y軸ラベル
    fontproperties=fp   # 使うフォント情報
)

# グリッドを表示
plt.grid(axis='both')

# 凡例表示
plt.legend(loc='lower left', prop=fp)

# 表示
plt.show()

先頭と途中と最後を加筆。

#!/usr/bin/env python
# coding: utf-8

import numpy as np
import matplotlib as mpl
mpl.use('Agg')
import matplotlib.pyplot as plt
from matplotlib.font_manager import FontProperties

fname_result = 'result.txt'
fname_work = 'work.txt'


def score(fname):
    '''結果ファイルからスコア算出
    結果ファイルを読み込んで、正解率、適合率、再現率、F1スコアを返す

    戻り値：
    正解率,適合率,再現率,F1スコア
    '''
    # 結果を読み込んで集計
    TP = 0      # True-Positive     予想が+1、正解も+1
    FP = 0      # False-Positive    予想が+1、正解は-1
    FN = 0      # False-Negative    予想が-1、正解は+1
    TN = 0      # True-Negative     予想が-1、正解も-1

    with open(fname) as data_file:
        for line in data_file:
            cols = line.split('\t')

            if len(cols) < 3:
                continue

            if cols[0] == '+1':         # 正解
                if cols[1] == '+1':     # 予想
                    TP += 1
                else:
                    FN += 1
            else:
                if cols[1] == '+1':
                    FP += 1
                else:
                    TN += 1

    # 算出
    accuracy = (TP + TN) / (TP + FP + FN + TN)      # 正解率
    precision = TP / (TP + FP)      # 適合率
    recall = TP / (TP + FN)     # 再現率
    f1 = (2 * recall * precision) / (recall + precision)    # F1スコア

    return accuracy, precision, recall, f1


# 結果読み込み、予測確率は元の値（仮説関数hypothesis()の値）に戻す
results = []
with open(fname_result) as data_file:
    for line in data_file:

        cols = line.split('\t')
        if len(cols) < 3:
            continue

        # 正解ラベル
        label = cols[0]

        # 識別関数predict()の値
        if cols[1] == '-1':
            predict = 1.0 - float(cols[2])      # 確率を戻す
        else:
            predict = float(cols[2])

        results.append((label, predict))

# 閾値を変えながらスコア算出、グラフ描画用の配列へセット
fig = plt.figure()

thresholds = []
accuracys = []
precisions = []
recalls = []
f1s = []
for threshold in np.arange(0.02, 1.0, 0.02):

    # score()を使うため、一時ファイルに結果保存
    with open(fname_work, 'w') as file_out:
        for label, predict in results:
            if predict > threshold:
                file_out.write('{}\t{}\t{}\n'.format(label, '+1', predict))
            else:
                file_out.write('{}\t{}\t{}\n'.format(label, '-1', 1 - predict))

    # スコア算出
    accuracy, precision, recall, f1 = score(fname_work)

    # 結果追加
    thresholds.append(threshold)
    accuracys.append(accuracy)
    precisions.append(precision)
    recalls.append(recall)
    f1s.append(f1)


# グラフで使うフォント情報(デフォルトのままでは日本語が表示できない)
fp = FontProperties(
    fname='/Library/Fonts/Times New Roman Bold Italic.ttf'
)

# 折線グラフの値の設定
plt.plot(thresholds, accuracys, color='green', linestyle='--', label='正解率')
plt.plot(thresholds, precisions, color='red', linewidth=3, label='適合率')
plt.plot(thresholds, recalls, color='blue', linewidth=3, label='再現率')
plt.plot(thresholds, f1s, color='magenta', linestyle='--', label='F1スコア')

# 軸の値の範囲の調整
plt.xlim(
    xmin=0, xmax=1.0
)
plt.ylim(
    ymin=0, ymax=1.0
)

# グラフのタイトル、ラベル指定
plt.title(
    '79. 適合率-再現率グラフの描画',    # タイトル
    fontproperties=fp   # 使うフォント情報
)
plt.xlabel(
    'ロジスティック回帰モデルの分類の閾値',       # x軸ラベル
    fontproperties=fp   # 使うフォント情報
)
plt.ylabel(
    '精度',         # y軸ラベル
    fontproperties=fp   # 使うフォント情報
)

# グリッドを表示
plt.grid(axis='both')

# 凡例表示
plt.legend(loc='lower left', prop=fp)

# 表示
#plt.show()
fig.savefig('p79.png')

# ./p79.py
Traceback (most recent call last):
  File "./p79.py", line 144, in <module>
    fig.savefig('p79.png')
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/figure.py", line 2062, in savefig
    self.canvas.print_figure(fname, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backend_bases.py", line 2263, in print_figure
    **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py", line 517, in print_png
    FigureCanvasAgg.draw(self)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py", line 437, in draw
    self.figure.draw(self.renderer)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/artist.py", line 55, in draw_wrapper
    return draw(artist, renderer, *args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/figure.py", line 1493, in draw
    renderer, self, artists, self.suppressComposite)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/image.py", line 141, in _draw_list_compositing_images
    a.draw(renderer)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/artist.py", line 55, in draw_wrapper
    return draw(artist, renderer, *args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/axes/_base.py", line 2635, in draw
    mimage._draw_list_compositing_images(renderer, self, artists)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/image.py", line 141, in _draw_list_compositing_images
    a.draw(renderer)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/artist.py", line 55, in draw_wrapper
    return draw(artist, renderer, *args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/axis.py", line 1204, in draw
    self.label.draw(renderer)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/artist.py", line 55, in draw_wrapper
    return draw(artist, renderer, *args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/text.py", line 706, in draw
    bbox, info, descent = textobj._get_layout(renderer)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/text.py", line 300, in _get_layout
    ismath=False)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py", line 245, in get_text_width_height_descent
    font = self._get_agg_font(prop)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py", line 280, in _get_agg_font
    font = get_font(fname)
  File "/opt/conda/lib/python3.7/site-packages/matplotlib/font_manager.py", line 1389, in get_font
    return _get_font(filename, hinting_factor)
FileNotFoundError: [Errno 2] No such file or directory: '/Library/Fonts/Times New Roman Bold Italic.ttf'

fontを注釈にすると

# ./p79.py
Traceback (most recent call last):
  File "./p79.py", line 125, in <module>
    fontproperties=fp   # 使うフォント情報
NameError: name 'fp' is not defined

docker側のfont指定をしないといけないのかも。

最後までおよみいただきありがとうございました。

いいね　💚、フォローをお願いします。

Thank you very much for reading to the last sentence.

Please press the like icon 💚　and follow me for your happy life.