def test1():
""" Test to find F1 score by manually selecting product id(s) from original data to test """
script_dir = os.path.dirname(__file__)
file_to_learn = os.path.join(script_dir,
'../autofunc/assets/consumer_systems.csv')
train_data = make_df(file_to_learn)
# Use a threshold to get the top XX% of frequency values
threshold = 0.7
## Choose ID(s) from learning file to separate into the testing set
test_ids = [691, 169]
test_df, train_df = split_learning_verification(train_data, test_ids)
test_list = df_to_list(test_df)
comb_sort = counter_pandas(train_df)
thresh_results = get_top_results(comb_sort, threshold)
# Find the F1 score
learned_dict, matched, overmatched, unmatched, recall, precision, f1 = precision_recall(
thresh_results, test_list)
assert len(learned_dict) != 0
assert f1 > 0
def test_1():
"""
Testing that the highest confidence result for the screw component is couple solid, which is what
a screw does almost exclusively
"""
script_dir = os.path.dirname(__file__)
file_to_test = os.path.join(script_dir,
'../autofunc/assets/consumer_systems.csv')
test_data = pd.read_csv(file_to_test)
combos_sorted = counter_pandas(test_data)
assert combos_sorted['screw'][0][0] == 'couple solid'
def test_1():
""" Example showing how to automate functional representation with frequency values"""
# Dataset used for data mining
script_dir = os.path.dirname(__file__)
file_to_test = os.path.join(script_dir,
'../autofunc/assets/consumer_systems.csv')
test_data = pd.read_csv(file_to_test)
combos_sorted = counter_pandas(test_data)
# Use a threshold to get the top XX% of confidence values
threshold = 0.5
thresh_results = get_top_results(combos_sorted, threshold)
# Use a known product for verification
input_file = os.path.join(script_dir,
'../autofunc/assets/InputExample.csv')
# Get dictionary of functions and flows for each component based on data mining
results, unmatched = get_func_rep(thresh_results, input_file, True)
assert results['screw'][0][0] == 'couple solid'
assert 'cheese' in unmatched
from autofunc.get_top_results import get_top_results
from autofunc.counter_pandas import counter_pandas
from autofunc.get_precision_recall import precision_recall
from autofunc.df_to_list import df_to_list
import os.path
import pandas as pd
""" Example showing how to find F1 score using separate file of input components """
# Dataset used for data mining
script_dir = os.path.dirname(__file__)
file_to_learn = os.path.join(script_dir, '../autofunc/assets/consumer_systems.csv')
train_data = pd.read_csv(file_to_learn)
combos_sorted = counter_pandas(train_data)
# Use a threshold to get the top XX% of confidence values
threshold = 0.5
thresh_results = get_top_results(combos_sorted, threshold)
# Use a known product for verification
test_file = os.path.join(script_dir, '../autofunc/assets/jigsawQuery_headers.csv')
test_data = pd.read_csv(test_file)
test_list = df_to_list(test_data)
learned_dict, matched, overmatched, unmatched, recall, precision, f1 = precision_recall(thresh_results,
test_list)
# Optional write to file - uncomment and rename to write file
verification_ids = e
ver_df, learn_df = split_learning_verification(df, verification_ids)
ver_list = df_to_list(ver_df)
if not bd:
comb_sort, counts, combos = counter_pandas_with_counts(learn_df)
thresh_results = get_top_results(comb_sort, threshold)
# Find the F1 score of the verification test by comparing the learned results with the known function/flows
learned_dict, matched, overmatched, unmatched, recall, precision, f1 = precision_recall(
thresh_results, ver_list)
if bd:
bd_comb_sort = counter_pandas(bd_df)
bd_thresh_results = get_top_results(bd_comb_sort, threshold)
learned_dict, matched, overmatched, unmatched, recall, precision, f1 = precision_recall(
bd_thresh_results, ver_list)
precisions.append(precision)
recalls.append(recall)
print(e)
f1s += f1
keep.append([e, f1])
plots.append(f1)
avg_f1 = f1s / len(keep)
ps_thresh = i / 100
if reading:
keep_ids = similarity_df[
similarity_df[str(test_id)] > ps_thresh].index.tolist()
else:
keep_ids = similarity_df[
similarity_df[test_id] > ps_thresh].index.tolist()
keep_ids.remove(test_id)
# Only keep rows from data frame that have an id that is in the keep_ids list
keep_df = train_df[train_df['id'].isin(keep_ids)]
comb_sort = counter_pandas(keep_df)
# Component counting and fractions
train_comps = list(keep_df.comp.unique())
if train_comps:
comp_ratio = len(train_comps) / num_all_comps
comp_ratios.append((len(keep_ids), i, comp_ratio))
if comp_ratio > 0.7 and len(keep_ids) < 40:
keepers.append(keep_ids)
scatter_keep.append((comp_ratio, len(keep_ids)))
for t in range(10, 100, 5):
threshold = t / 100
ps_thresh = i / 100
if reading:
keep_ids = similarity_df[
similarity_df[str(test_id)] > ps_thresh].index.tolist()
else:
keep_ids = similarity_df[
similarity_df[test_id] > ps_thresh].index.tolist()
keep_ids.remove(test_id)
# Only keep rows from data frame that have an id that is in the keep_ids list
keep_df = train_df[train_df['id'].isin(keep_ids)]
comb_sort = counter_pandas(keep_df)
# Component counting and fractions
train_comps = list(keep_df.comp.unique())
if train_comps:
comp_ratios.append(
(len(keep_ids), i, len(train_comps) / num_all_comps))
for t in range(10, 100, 5):
threshold = t / 100
print(test_id, ' ', ps_thresh, ' ', threshold)
thresh_results = get_top_results(comb_sort, threshold)
if not keep_ids: