towhee.engine.graph_context.GraphContext¶

class towhee.engine.graph_context.GraphContext(ctx_idx: int, graph_repr: GraphRepr)[source]¶

Bases: object

GraphContext is a data processing network with one or multiple Operator. Each row of a Pipeline’s inputs will be processed individually by a GraphContext.

Parameters:

ctx_idx – (int) The index of this GraphContext.
graph_repr – (towhee.dag.GraphRepr) The DAG representation this GraphContext will implement.

Methods

`gc`
`join`
`result`
`slow_down`	Slow down the op whose df name it.
`speed_up`	spped up the op whose df name it.
`stop`

Attributes

`dataframes`
`inputs`	Returns the graph's input DataFrame.
`op_ctxs`
`outputs`	Returns the graph's output DataFrame.

__call__(inputs: Tuple)[source]¶: Call self as a function.

__init__(ctx_idx: int, graph_repr: GraphRepr)[source]¶

property inputs: DataFrame¶

Returns the graph’s input DataFrame.

Returns:

(towhee.dataframe.DataFrame): The input DataFrame.

property outputs: DataFrame¶

Returns the graph’s output DataFrame.

Returns:

(towhee.dataframe.DataFrame): The output DataFrame.

slow_down(df_name: str, time_sec: int)[source]¶: Slow down the op whose df name it.

speed_up(df_name: str)[source]¶: spped up the op whose df name it.