{% if messages %}
Warnings
{{ message.column_name }} has constant value "{{ message.values['mode'] }}"
|
Rejected | {% elif message.message_type == MessageType.CORR %}
{{ message.column_name }} is highly correlated with {{ message.values['correlation_var'] }} (ρ = {{ message.values['correlation'] | fmt_numeric }})
|
Rejected | {% elif message.message_type == MessageType.RECODED %}
{{ message.column_name }} is a recoding of {{ message.values['correlation_var'] }}
|
Rejected | {% elif message.message_type == MessageType.HIGH_CARDINALITY %}
{{ message.column_name }} has a high cardinality: {{ message.values['distinct_count'] }} distinct values
|
Warning | {% elif message.message_type == MessageType.UNSUPPORTED %}
{{ message.column_name }} is an unsupported type, check if it needs cleaning or further analysis
|
Warning | {% elif message.message_type == MessageType.DUPLICATES %}Dataset has {{ message.values['n_duplicates'] }} ({{ message.values['p_duplicates'] | fmt_percent }}) duplicate rows | Warning | {% elif message.message_type == MessageType.SKEWED %}
{{ message.column_name }} is highly skewed (γ1 = {{ message.values['skewness'] | fmt_numeric}})
|
Skewed | {% elif message.message_type == MessageType.MISSING %}
{{ message.column_name }} has {{ message.values['n_missing'] }} ({{ message.values['p_missing'] | fmt_percent }}) missing values
|
Missing | {% elif message.message_type == MessageType.INFINITE %}
{{ message.column_name }} has {{ message.values['n_infinite'] }} ({{ message.values['p_infinite'] | fmt_percent }}) infinite values
|
Infinite | {% elif message.message_type == MessageType.ZEROS %}
{{ message.column_name }} has {{ message.values['n_zeros'] }} ({{ message.values['p_zeros'] | fmt_percent }}) zeros
|
Zeros | {% elif message.message_type == MessageType.TYPE_DATE %}
{{ message.column_name }} only contains datetime values, but is categorical. Consider applying pd.to_datetime()
|
Type | {% else %}Unknown type {{ message['mtype'] }} | {% endif %}