# AggregateFilter **`class AggregateFilter(config)`** Bases: [`Filter`](squirro.lib.nlp.steps.filters.Filter.md#squirro.lib.nlp.steps.filters.Filter) The aggregate [`Filter`](squirro.lib.nlp.steps.filters.Filter.md#squirro.lib.nlp.steps.filters.Filter) concatenates / sums up the field values of the aggregated_fields grouped by the value of the aggregating_field **Input** - the aggregated_fields need to be of type [`int`](https://docs.python.org/3.11/library/functions.html#int), [`float`](https://docs.python.org/3.11/library/functions.html#float) or [`list`](https://docs.python.org/3.11/library/stdtypes.html#list) **Output** - the output field is filled with data of type [`int`](https://docs.python.org/3.11/library/functions.html#int), [`float`](https://docs.python.org/3.11/library/functions.html#float) or [`list`](https://docs.python.org/3.11/library/stdtypes.html#list) according to the type of the aggregated_fields Parameters - `type` ([`str`](https://docs.python.org/3.11/library/stdtypes.html#str)) – aggregate - `aggregated_fields` ([`list`](https://docs.python.org/3.11/library/stdtypes.html#list)) – Fields to aggregate - `aggregating_field` ([`str`](https://docs.python.org/3.11/library/stdtypes.html#str)) – Field to aggregate into - `output_field` ([`str`](https://docs.python.org/3.11/library/stdtypes.html#str)`, ``None`) – Field to store the aggregation (defaults to aggregating_field) - `use_separate_docs` ([`bool`](https://docs.python.org/3.11/library/functions.html#bool)`, ``True`) – Whether or not to use a separate doc for each aggregating value **Example** ```json { "step": "filter", "type": "aggregate", "output_field": "out", "aggregated_fields": ["b", "c"], "aggregating_field": "a", } ``` Methods SummaryMethods Documentation **`process(docs)`** Process a set of documents Parameters `docs` (`generator``(`[`Document`](../technical/libnlp/base.md#squirro.lib.nlp.document.Document)`)`) – Generator of documents Returns Generator of processed documents Return type generator([Document](../technical/libnlp/base.md#squirro.lib.nlp.document.Document)) **`train(docs)`** Train on a step of a set of documents Parameters `docs` (`generator``(`[`Document`](../technical/libnlp/base.md#squirro.lib.nlp.document.Document)`)`) – Generator of documents Returns Generator of processed documents Return type generator([Document](../technical/libnlp/base.md#squirro.lib.nlp.document.Document))