Python Workflow Examples

Example #1

def main():

    filter_options = Options.crystal_progression
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getCrystalDecJanLogDF

    w = Workflow(filter_options=filter_options, nested_folder_output=False)
    w.pca_dimension_count = 2
    w.clustering_counts = [4]
    w.RunWorkflow(get_df_func=df_getter)

Example #2

def main():
    filter_options = Options.crystal_feedback
    df_getter = cu.getCrystalDecJanLogDF

    w = Workflow(filter_options=filter_options, nested_folder_output=False)
    w.pca_dimension_count = 2
    # w.clustering_counts = range(3, 8)
    w.clustering_counts = [6]
    w.verbose = True
    w.RunWorkflow(get_df_func=df_getter)

Example #3

File: john_crystal_actions.py Project: fielddaylab/chi_play_clustering

def main():
    filter_options = Options.crystal_actions
    df_getter = cu.getCrystalDecJanLogDF

    w = Workflow(filter_options=filter_options, nested_folder_output=False)
    w.pca_dimension_count = 2
    w.further_filter_query_list = [
        f'sum_lvl_0_to_4_avgMoleculeDragDurationInSecs < {5*60}'
    ]  # 5 mins - one person averaged 8+ minutes
    w.clustering_counts = [4]
    w.RunWorkflow(get_df_func=df_getter)

Example #4

File: john_luke_waves_actions.py Project: fielddaylab/chi_play_clustering

from src import settings
from src.cluster_workflow import Workflow
from src import cluster_utils as cu
from src.options import Options

if __name__ == '__main__':
    # import setup
    # setup.init_path()

    filter_options = Options.waves_actions_lv016
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getWavesDecJanLogDF

    w = Workflow(filter_options=filter_options)
    w.pca_dimension_count = 2
    for k in range(4, 8):
        w.clustering_count = k
        w.RunWorkflow(get_df_func=df_getter)

Example #5

File: john_waves_progression.py Project: fielddaylab/chi_play_clustering

from src import settings
from src.cluster_workflow import Workflow
from src import cluster_utils as cu
from src.options import Options

if __name__ == '__main__':
    # import setup
    # setup.init_path()

    filter_options = Options.waves_progression
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getWavesDecJanLogDF

    w = Workflow(filter_options=filter_options)
    w.further_filter_query_list = [
        f'sum_lvl_0_to_34_totalLevelTime < {50*60}'
    ]  # 50 mins
    w.clustering_counts = range(3, 8)
    w.verbose = True
    w.RunWorkflow(get_df_func=df_getter)

Example #6

from src import settings
from src.cluster_workflow import Workflow
from src import cluster_utils as cu
from src.options import Options

if __name__ == '__main__':
    # import setup
    # setup.init_path()

    filter_options = Options.lakeland_actions_lvl01
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getLakelandDecJanLogDF

    w = Workflow(filter_options=filter_options, nested_folder_output=False)
    w.pca_dimension_count = 2
    w.clustering_counts = [6]
    w.RunWorkflow(get_df_func=df_getter)

Example #7

File: john_waves_feedback.py Project: fielddaylab/chi_play_clustering

from src import settings
from src.cluster_workflow import Workflow
from src import cluster_utils as cu
from src.options import Options

if __name__ == '__main__':
    # import setup
    # setup.init_path()

    filter_options = Options.waves_feedback_lv016
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getWavesDecJanLogDF

    w = Workflow(filter_options=filter_options)
    w.do_logtransform = False
    w.pca_dimension_count = 2
    w.clustering_counts = range(3, 8)
    w.verbose = True
    w.RunWorkflow(get_df_func=df_getter)

Example #8

from src import settings
from src.cluster_workflow import Workflow
from src import cluster_utils as cu
from src.options import Options

if __name__ == '__main__':
    # import setup
    # setup.init_path()

    filter_options = Options.lakeland_feedback_lv01_with_bloom
    output_foler = settings.OUTPUT_DIR
    df_getter = cu.getLakelandDecJanLogDF

    w = Workflow(filter_options=filter_options, nested_folder_output=False)
    w.clustering_method = "KMeans"
    w.pca_dimension_count = 2
    w.eps_min_list = [(eps, min_samples)
                      for eps in [.01, .02, .05, .07, .1, .2, .3]
                      for min_samples in [5]]
    w.min_cluster_size_list = [15, 30, 60, 100]
    w.plot_silhouettes = True
    w.plot_radars = True
    w.clustering_counts = [7]
    w.RunWorkflow(get_df_func=df_getter)