ExecutionContext

from AID.ctx.interface import ExecutionContext

Execution Context — Pipeline Orchestration & Phase Execution. The primary API surface for executing the AIDflow pipeline. Provides phase-based access to all workers and a unified process method for sequential execution. Executes workers by domain-specific pipeline phase: data - encoding - training - selection.

Automated execution of all pipeline steps or a selected phase via ctx.flow.process()
Execution runs within a defined context scope (ctx.setup), supplied by ProjectContext
Pipeline steps are sequence-dependent - e.g., the encoding phase cannot be executed prior to the data phase

1. data
────────────────────

data_loader
fe_cleaner
fe_stabilizer
obs_stabilizer

2. encoding
────────────────────

num_encoder
cat_encoder
ixn_encoder
woe_encoder
woe_stabilizer
data_encoder

3. training
────────────────────

agent1_predictor
agent2_predictor
agent3_predictor
agent4_predictor

4. selection
────────────────────

agents_predictor
agents_comparator
agents_transition
agent_selector
champion_predictor

Examples

Defining the Project Context (Control Plane) & Project Scope:

import aidflow as aid

ctx = aid.ProjectContext()

ctx.genesis(
    project      = 'test',
    input_data   = 'test.csv',
    target       = 'target',
    yy_mm        = 'decision_date',
    yy_mm_format = '%Y-%m-%d'
)
ctx.params()

Run the whole flow within the defined context:

ctx.flow.process('all', verbose = 0)

Run the whole flow within the defined context with custom data present in memory:

import pandas as pd

my_data = pd.DataFrame(dict(
    target = [...], 
    decision_data = [...],
    ...
))

ctx.flow.process('all', data = my_data, verbose = 1)

Run selected flow phase within the defined context:

ctx.flow.process('data', verbose = 2)

Run selected single worker within the defined context:

data_loader = ctx.flow.data.data_loader()
...
ixn_encoder = ctx.flow.encoding.ixn_encoder()
...
agent2_predictor = ctx.flow.training.agent2_predictor()
...
agent_selector = ctx.flow.selection.agent_selector()

Execution Interface

Data Interface

Encoding Interface

Training Interface

Selection Interface

Inference Interface

Examples

​Examples

Examples