如何将数据集A列值匹配对应Emb_A结果并新增列？

Got it, let's walk through how to add the Emb_A column to your dataset by mapping values from column A to their corresponding embeddings. Here's a straightforward solution using pandas, which is perfect for this kind of data matching task:

Step 1: Set up your data and mapping #

First, we'll load your original dataset and create a dictionary for the A-to-Emb_A mapping. This dictionary makes lookups fast and easy.

import pandas as pd

# Load your original dataset
original_df = pd.DataFrame({
    'id': [1, 1, 1, 1, 1, 1, 1, 1, 1],
    'tslot': [
        '2014-11-02 22:45:00',
        '2014-10-26 09:15:00',
        '2014-10-26 11:00:00',
        '2014-10-26 11:15:00',
        '2014-10-26 12:00:00',
        '2014-10-26 13:00:00',
        '2014-10-26 13:45:00',
        '2014-10-26 14:45:00',
        '2014-10-26 15:45:00'
    ],
    'A': [89, 762, 25, 762, 25, 25, 25, 902, 902]
})

# Create the A-to-Emb_A mapping dictionary
embedding_map = {
    762.0: -0.237305,
    89.0: -0.033929,
    902.0: -0.237330,
    25.0: -0.237300,
    1176.0: -0.237300,
    827.0: -0.237312,
    1227.0: -0.237301
}

Step 2: Match data types and add the Emb_A column #

Notice that your mapping uses float keys (like 762.0) but the original A column has integers. We'll convert A to float to ensure matching works correctly, then use pandas' map() function to assign the corresponding Emb_A values.

# Convert column A to float to match the mapping's key type
original_df['A'] = original_df['A'].astype(float)

# Add the Emb_A column by mapping values in A
original_df['Emb_A'] = original_df['A'].map(embedding_map)

Step 3: Check the result #

If you print the dataframe now, you'll get exactly the output you're looking for:

print(original_df)

Output:

id                tslot      A    Emb_A
0   1  2014-11-02 22:45:00   89.0 -0.033929
1   1  2014-10-26 09:15:00  762.0 -0.237305
2   1  2014-10-26 11:00:00   25.0 -0.237300
3   1  2014-10-26 11:15:00  762.0 -0.237305
4   1  2014-10-26 12:00:00   25.0 -0.237300
5   1  2014-10-26 13:00:00   25.0 -0.237300
6   1  2014-10-26 13:45:00   25.0 -0.237300
7   1  2014-10-26 14:45:00  902.0 -0.237330
8   1  2014-10-26 15:45:00  902.0 -0.237330

Bonus: Handle missing values #

If there are values in A that aren't in your mapping, map() will return NaN for those rows. You can fill those with a default value (like 0.0) using fillna():

original_df['Emb_A'] = original_df['A'].map(embedding_map).fillna(0.0)

内容的提问来源于stack exchange，提问作者Krush23