adding pyabc

app/flows/run_abc.py

@@ -9,11 +9,12 @@
 import numpy as np
 import pandas as pd
 import plotly.express as px
+import pymc as pm
 from joblib import dump as calibrator_dump
 from matplotlib import pyplot as plt
 from prefect import flow, task
 from prefect.artifacts import create_table_artifact
-from prefect.task_runners import ConcurrentTaskRunner
+from prefect.task_runners import SequentialTaskRunner
 
 from deeprootgen.calibration import (
     SensitivityAnalysisModel,
@@ -42,6 +43,11 @@
 ######################################
 
 
+def distance_func(e: float, observed: np.ndarray, simulated: np.ndarray) -> float:
+    print(simulated)
+    return simulated.item()
+
+
 @task
 def run_abc(input_parameters: RootCalibrationModel, simulation_uuid: str) -> None:
     """Running Approximate Bayesian Computation.
@@ -55,6 +61,9 @@ def run_abc(input_parameters: RootCalibrationModel, simulation_uuid: str) -> Non
     begin_experiment(TASK, simulation_uuid, input_parameters.simulation_tag)
     log_experiment_details(simulation_uuid)
 
+    # distance, statistics_list = get_calibration_summary_stats(input_parameters)
+    # [statistic.statistic_value for statistic in statistics_list]
+
     config = input_parameters.dict()
     log_config(config, TASK)
     mlflow.end_run()
@@ -63,7 +72,7 @@ def run_abc(input_parameters: RootCalibrationModel, simulation_uuid: str) -> Non
 @flow(
     name="abc",
     description="Perform Bayesian parameter estimation for the root model using Approximate Bayesian Computation.",
-    task_runner=ConcurrentTaskRunner(),
+    task_runner=SequentialTaskRunner(),
 )
 def run_abc_flow(input_parameters: RootCalibrationModel, simulation_uuid: str) -> None:
     """Flow for running Approximate Bayesian Computation.
@@ -75,3 +84,76 @@ def run_abc_flow(input_parameters: RootCalibrationModel, simulation_uuid: str) -
             The simulation uuid.
     """
     run_abc.submit(input_parameters, simulation_uuid)
+
+    # names = []
+    # priors = []
+    # with pm.Model() as model:
+    #     parameter_intervals = input_parameters.parameter_intervals.dict()
+    #     for name, v in parameter_intervals.items():
+    #         names.append(name)
+
+    #         lower_bound = v["lower_bound"]
+    #         upper_bound = v["upper_bound"]
+    #         data_type = v["data_type"]
+
+    #         if data_type == "discrete":
+    #             prior = pm.DiscreteUniform(name, lower_bound, upper_bound)
+    #         else:
+    #             prior = pm.Uniform(name, lower_bound, upper_bound)
+    #         priors.append(prior)
+
+    #     params = tuple(priors)
+
+    #     def simulator_func(
+    #         _, *parameters
+    #     ):
+    #         parameters = parameters[:-1]
+
+    #         parameter_specs = {}
+    #         for i, name in enumerate(names):
+    #             parameter_specs[name] = parameters[i].item()
+
+    #         discrepancy = calculate_summary_statistic_discrepancy(
+    #             parameter_specs, input_parameters, statistics_list, distance
+    #         )
+
+    #         print(parameter_specs)
+    #         print(discrepancy)
+    #         return np.array([discrepancy])
+
+    #     calibration_parameters = input_parameters.calibration_parameters
+    #     pm.Simulator(
+    #         "root_simulator",
+    #         simulator_func,
+    #         params = params,
+    #         distance = distance_func,
+    #         sum_stat = "identity",
+    #         epsilon = calibration_parameters["epsilon"],
+    #         observed = observed_values,
+    #     )
+
+    #     time_now = get_datetime_now()
+    #     outdir = get_outdir()
+    #     pgm = pm.model_to_graphviz(model = model)
+    #     outfile =  f"{time_now}-{TASK}_model_graph"
+    #     pgm.render(format = "png", directory = outdir, filename = outfile)
+    #     outfile = osp.join(outdir, f"{outfile}.png")
+    #     mlflow.log_artifact(outfile)
+
+    # trace = pm.sample_smc(
+    #     # draws = calibration_parameters["draws"],
+    #     model = model,
+    #     draws = 3,
+    #     chains = 1,
+    #     cores = 1,
+    #     # chains = calibration_parameters["chains"],
+    #     # cores = calibration_parameters["cores"],
+    #     # compute_convergence_checks = False,
+    #     # return_inferencedata = False,
+    #     # random_seed = input_parameters.random_seed,
+    #     # progressbar = False
+    # )
+
+    # del trace
+    # del model
+    # print(trace)

app/pages/abc_root_system.py

@@ -359,13 +359,21 @@ def toggle_calibration_parameters_collapse(n: int, is_open: bool) -> bool:
     Output({"index": f"{PAGE_ID}-run-sim-button", "type": ALL}, "disabled"),
     Output({"index": f"{PAGE_ID}-clear-obs-data-file-button", "type": ALL}, "disabled"),
     Input("store-summary-data", "data"),
+    Input({"index": f"{PAGE_ID}-select-summary-stats-dropdown", "type": ALL}, "value"),
+    Input({"index": f"{PAGE_ID}-distance-dropdown", "type": ALL}, "value"),
 )
-def update_summary_data_state(summary_data: dict | None) -> tuple:
+def update_summary_data_state(
+    summary_data: dict | None, summary_stats: list, distances: list
+) -> tuple:
     """Update the state of the summary data.
 
     Args:
         summary_data (dict | None):
             The summary data.
+        summary_stats (list):
+            The list of summary statistics.
+        distances (list):
+            The list of distance metrics.
 
     Returns:
         tuple:
@@ -379,6 +387,17 @@ def update_summary_data_state(summary_data: dict | None) -> tuple:
     if summary_label is None:
         return button_contents, [True], [True]
 
+    if summary_stats is None or distances is None:
+        return [summary_label], [True], [True]
+
+    summary_stats_list = summary_stats[0]
+    distance_list = distances[0]
+    if len(summary_stats_list) == 0 or distance_list is None:
+        return [summary_label], [True], [True]
+
+    if summary_stats_list[0] is None or distance_list == "":
+        return [summary_label], [True], [True]
+
     return [summary_label], [False], [False]
 
 

app/pages/optimisation_root_system.py

@@ -359,13 +359,21 @@ def toggle_calibration_parameters_collapse(n: int, is_open: bool) -> bool:
     Output({"index": f"{PAGE_ID}-run-sim-button", "type": ALL}, "disabled"),
     Output({"index": f"{PAGE_ID}-clear-obs-data-file-button", "type": ALL}, "disabled"),
     Input("store-summary-data", "data"),
+    Input({"index": f"{PAGE_ID}-select-summary-stats-dropdown", "type": ALL}, "value"),
+    Input({"index": f"{PAGE_ID}-distance-dropdown", "type": ALL}, "value"),
 )
-def update_summary_data_state(summary_data: dict | None) -> tuple:
+def update_summary_data_state(
+    summary_data: dict | None, summary_stats: list, distances: list
+) -> tuple:
     """Update the state of the summary data.
 
     Args:
         summary_data (dict | None):
             The summary data.
+        summary_stats (list):
+            The list of summary statistics.
+        distances (list):
+            The list of distance metrics.
 
     Returns:
         tuple:
@@ -379,6 +387,17 @@ def update_summary_data_state(summary_data: dict | None) -> tuple:
     if summary_label is None:
         return button_contents, [True], [True]
 
+    if summary_stats is None or distances is None:
+        return [summary_label], [True], [True]
+
+    summary_stats_list = summary_stats[0]
+    distance_list = distances[0]
+    if len(summary_stats_list) == 0 or distance_list is None:
+        return [summary_label], [True], [True]
+
+    if summary_stats_list[0] is None or distance_list == "":
+        return [summary_label], [True], [True]
+
     return [summary_label], [False], [False]
 
 

app/pages/sensitivity_analysis_root_system.py

@@ -359,13 +359,21 @@ def toggle_calibration_parameters_collapse(n: int, is_open: bool) -> bool:
     Output({"index": f"{PAGE_ID}-run-sim-button", "type": ALL}, "disabled"),
     Output({"index": f"{PAGE_ID}-clear-obs-data-file-button", "type": ALL}, "disabled"),
     Input("store-summary-data", "data"),
+    Input({"index": f"{PAGE_ID}-select-summary-stats-dropdown", "type": ALL}, "value"),
+    Input({"index": f"{PAGE_ID}-distance-dropdown", "type": ALL}, "value"),
 )
-def update_summary_data_state(summary_data: dict | None) -> tuple:
+def update_summary_data_state(
+    summary_data: dict | None, summary_stats: list, distances: list
+) -> tuple:
     """Update the state of the summary data.
 
     Args:
         summary_data (dict | None):
             The summary data.
+        summary_stats (list):
+            The list of summary statistics.
+        distances (list):
+            The list of distance metrics.
 
     Returns:
         tuple:
@@ -379,6 +387,17 @@ def update_summary_data_state(summary_data: dict | None) -> tuple:
     if summary_label is None:
         return button_contents, [True], [True]
 
+    if summary_stats is None or distances is None:
+        return [summary_label], [True], [True]
+
+    summary_stats_list = summary_stats[0]
+    distance_list = distances[0]
+    if len(summary_stats_list) == 0 or distance_list is None:
+        return [summary_label], [True], [True]
+
+    if summary_stats_list[0] is None or distance_list == "":
+        return [summary_label], [True], [True]
+
     return [summary_label], [False], [False]
 
 

pyproject.toml

@@ -54,6 +54,8 @@ dash-daq = "^0.5.0"
 ydata-profiling = "^4.10.0"
 adbnx-adapter = "^5.0.3"
 python-arango = "<8.0"
+pyabc = "^0.12.13"
+redis = "^5.0.8"
 
 
 [tool.poetry.group.torch.dependencies]