Question

我尝试根据新列中的文本在pandas数据框中添加新列，例如这是我的数据：

 >>> data

 No    Description
 1     Extention Slack 1 Month
 2     Extention Slack 1 Year
 3     Slack 6 Month
 4     Slack 1 Year

我需要的是

 No    Description                 M    M+1   M+2  M+3  M+4   M+5  M+6 ... M+11
 1     Extention Slack 1 Month    1    0     0    0    0     0    0       0
 2     Extention Slack 1 Year     1    1     1    1    1     1    1       1
 3     Slack 6 Month              1    1     1    1    1     1    0       0
 4     Slack 3 Month              1    1     1    0    0     0    0       0

我做的是

import numpy as np
data['M'] = np.where(data['Description'].str.contains('1 Year'), 1, 0)

我猜怎么做？

Answer 1

在“说明”列中，您希望根据def train(X_train, y_train, X_val, y_val, batch_size, iterations=1000, learning_rate=0.5): x = tf.placeholder(tf.float32, [None, 784]) w1 = tf.Variable(tf.zeros([784, 100])) b1 = tf.Variable(tf.zeros([100])) #a1 = tf.placeholder(tf.float32, [None, 100]) w2 = tf.Variable(tf.zeros([100, 11])) b2 = tf.Variable(tf.zeros([11])) y_ = tf.placeholder(tf.float32, [None, 11]) z1 = tf.matmul(x, w1) + b1 a1 = tf.nn.relu(z1) y = tf.matmul(a1, w2) + b2 cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y)) train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(cross_entropy) correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) sess = tf.InteractiveSession() tf.global_variables_initializer().run() #train parameters = {} i = 0 j = batch_size for k in range(iterations): batch_x = X_train[i:j,:] batch_y = y_train[i:j,:] print batch_x.shape print batch_y.shape i += batch_size j += batch_size if(i > X_train.shape[0] or j > y_train.shape[0]): i += batch_size % X_train.shape[0] j += batch_size % y_train.shape[0] sess.run(train_step, feed_dict={x: batch_x, y_: batch_y}) if( k % 100 == 0): print(sess.run(accuracy, feed_dict={x: X_val, y_: y_val})) parameters = {"w1": w1, "b1": b1, "w2": w2, "b2": b2 } return parameters或{time} {time_label}部分推断，在12个月内填充一个或零。

这是一种做你想做的事情的方法：

1 Year

完全可重复的例子：

1 Month

输出：

# create two temporary columns 
# time: holds the numeric value associated with time_label (month or year)
df['time'], df['time_label'] = df.Description.str.split().apply(lambda x: pd.Series(x[-2:])).values.T

# define the numeric equivalent of Month and Year 
mapping = {"Month":1, "Year":12}

for month in range(12):
    # if is only here to pretty print M, M+1, M+2, ...
    # you can remove it if you accept M+0, M+1, ...  
    if month == 0:
        df["M"] = np.where(df.time.astype(int)*df.time_label.map(mapping) >= month+1, 1, 0)
    else:
        df["M"+"+"+str(month)] = np.where(df.time.astype(int)*df.time_label.map(mapping) >= month+1, 1, 0)

如何在pandas dataframe中组合条件格式和str.contains来创建新列？

1 个答案: