0
def prep_data(embeddings, df):
    """
    Prepares training and test set the train and test data used.

    :return: train and test set
    """
    #Original_DF
    OD_df = df.reset_index()
    OD_df.loc[:,'type'] = 'nails'
    
    embeddings_df = pd.DataFrame(data=embeddings)
    embeddings_df = embeddings_df.reset_index()
    
    embedded_df = pd.merge(embeddings_df, OD_df, on='index')
    
    train, valid = train_test_split(embedded_df, test_size=0.2, random_state=42, shuffle=True)
    #train_col = ['train'] * len(train) 
    #train['split']= train_col 

    train.loc[:,'split'] = 'train'
    
    #valid_col = ['valid'] * len(valid) 
    #valid['split']= valid_col 
    valid.loc[:,'split'] = 'valid'
    
    x_train, y_train = train.iloc[:, :3840], train.loc[:,'nails']
    x_valid, y_valid = valid.iloc[:, :3840], valid.loc[:,'nails']


    return x_train, y_train, x_valid, y_valid

Calling Code

x_train, y_train, x_valid, y_valid = prep_data(embeddings, df)

I get the following warning. I've tried implemented the .iloc method and .loc as well, even just normal subsetting without the previous but it continues to give me the error message. When I revise the dataframes, everything seems to be in order, but I'm not sure on how to suppress or overcome this warning.

warning

/home/maria/my_python_env/lib/python3.7/site-packages/pandas/core/indexing.py:1596: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
mkrieger1
  • 19,194
  • 5
  • 54
  • 65
bioinformatics_student
  • 488
  • 1
  • 8
  • 21
  • Did you try: `pd.options.mode.chained_assignment = None` to suppress warning? https://stackoverflow.com/a/20627316/5916727 – niraj Sep 24 '20 at 15:34
  • This is a warning message. Not an error. – gtomer Sep 24 '20 at 15:43
  • 1
    Generally more reasonable to understand the cause and address the warning than to simply suppress it. The warning is there for a reason. Yes, it’s a book in itself, but [this post](https://stackoverflow.com/a/53954986/6340496) explains it nicely. – S3DEV Sep 24 '20 at 17:22

0 Answers0