In [1]:
from popsborder.scenarios import run_scenarios, load_scenario_table
from popsborder.simulation import load_configuration
from popsborder.outputs import save_scenario_result_to_pandas
In [2]:
import pandas as pd
import re
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline
In [3]:
from pathlib import Path
datadir = Path("tests/test_summaries")
basic_config = load_configuration(datadir / "config.yml")
scenario_table = load_scenario_table(datadir / "scenarios_config.csv")
In [4]:
results = run_scenarios(
config=basic_config,
scenario_table=scenario_table,
seed=42,
num_simulations=2,
num_consignments=10,
)
Running scenario: clustered_10 Running scenario: clustered_12 Running scenario: clustered_14 Running scenario: clustered_16 Running scenario: clustered_09 Running scenario: clustered_11 Running scenario: clustered_13 Running scenario: clustered_15 Running scenario: clustered_02 Running scenario: clustered_04 Running scenario: clustered_06 Running scenario: clustered_08 Running scenario: clustered_01 Running scenario: clustered_03 Running scenario: clustered_05 Running scenario: clustered_07 Running scenario: clustered_35 Running scenario: clustered_36 Running scenario: clustered_39 Running scenario: clustered_40 Running scenario: clustered_43 Running scenario: clustered_44 Running scenario: clustered_47 Running scenario: clustered_48 Running scenario: clustered_34 Running scenario: clustered_38 Running scenario: clustered_42 Running scenario: clustered_46 Running scenario: clustered_33 Running scenario: clustered_37 Running scenario: clustered_41 Running scenario: clustered_45 Running scenario: clustered_19 Running scenario: clustered_20 Running scenario: clustered_23 Running scenario: clustered_24 Running scenario: clustered_27 Running scenario: clustered_28 Running scenario: clustered_31 Running scenario: clustered_32 Running scenario: clustered_18 Running scenario: clustered_22 Running scenario: clustered_26 Running scenario: clustered_30 Running scenario: clustered_17 Running scenario: clustered_21 Running scenario: clustered_25 Running scenario: clustered_29 Running scenario: clustered_large_10 Running scenario: clustered_large_12 Running scenario: clustered_large_14 Running scenario: clustered_large_16 Running scenario: clustered_large_09 Running scenario: clustered_large_11 Running scenario: clustered_large_13 Running scenario: clustered_large_15 Running scenario: clustered_large_02 Running scenario: clustered_large_04 Running scenario: clustered_large_06 Running scenario: clustered_large_08 Running scenario: clustered_large_01 Running scenario: clustered_large_03 Running scenario: clustered_large_05 Running scenario: clustered_large_07 Running scenario: clustered_large_35 Running scenario: clustered_large_36 Running scenario: clustered_large_39 Running scenario: clustered_large_40 Running scenario: clustered_large_43 Running scenario: clustered_large_44 Running scenario: clustered_large_47 Running scenario: clustered_large_48 Running scenario: clustered_large_34 Running scenario: clustered_large_38 Running scenario: clustered_large_42 Running scenario: clustered_large_46 Running scenario: clustered_large_33 Running scenario: clustered_large_37 Running scenario: clustered_large_41 Running scenario: clustered_large_45 Running scenario: clustered_large_19 Running scenario: clustered_large_20 Running scenario: clustered_large_23 Running scenario: clustered_large_24 Running scenario: clustered_large_27 Running scenario: clustered_large_28 Running scenario: clustered_large_31 Running scenario: clustered_large_32 Running scenario: clustered_large_18 Running scenario: clustered_large_22 Running scenario: clustered_large_26 Running scenario: clustered_large_30 Running scenario: clustered_large_17 Running scenario: clustered_large_21 Running scenario: clustered_large_25 Running scenario: clustered_large_29 Running scenario: clustered_small_10 Running scenario: clustered_small_12 Running scenario: clustered_small_14 Running scenario: clustered_small_16 Running scenario: clustered_small_09 Running scenario: clustered_small_11 Running scenario: clustered_small_13 Running scenario: clustered_small_15 Running scenario: clustered_small_02 Running scenario: clustered_small_04 Running scenario: clustered_small_06 Running scenario: clustered_small_08 Running scenario: clustered_small_01 Running scenario: clustered_small_03 Running scenario: clustered_small_05 Running scenario: clustered_small_07 Running scenario: clustered_small_35 Running scenario: clustered_small_36 Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Running scenario: clustered_small_39 Running scenario: clustered_small_40 Running scenario: clustered_small_43 Running scenario: clustered_small_44 Running scenario: clustered_small_47 Running scenario: clustered_small_48 Running scenario: clustered_small_34 Running scenario: clustered_small_38 Running scenario: clustered_small_42 Running scenario: clustered_small_46 Running scenario: clustered_small_33 Running scenario: clustered_small_37 Running scenario: clustered_small_41 Running scenario: clustered_small_45 Running scenario: clustered_small_19 Running scenario: clustered_small_20 Running scenario: clustered_small_23 Running scenario: clustered_small_24 Running scenario: clustered_small_27 Running scenario: clustered_small_28 Running scenario: clustered_small_31 Running scenario: clustered_small_32 Running scenario: clustered_small_18 Running scenario: clustered_small_22 Running scenario: clustered_small_26 Running scenario: clustered_small_30 Running scenario: clustered_small_17 Running scenario: clustered_small_21 Running scenario: clustered_small_25 Running scenario: clustered_small_29 Running scenario: contamination_10_10 Running scenario: contamination_10_12 Running scenario: contamination_10_14 Running scenario: contamination_10_16 Running scenario: contamination_10_09 Running scenario: contamination_10_11 Running scenario: contamination_10_13 Running scenario: contamination_10_15 Running scenario: contamination_10_02 Running scenario: contamination_10_04 Running scenario: contamination_10_06 Running scenario: contamination_10_08 Running scenario: contamination_10_01 Running scenario: contamination_10_03 Running scenario: contamination_10_05 Running scenario: contamination_10_07 Running scenario: contamination_10_35 Running scenario: contamination_10_36 Running scenario: contamination_10_39 Running scenario: contamination_10_40 Running scenario: contamination_10_43 Running scenario: contamination_10_44 Running scenario: contamination_10_47 Running scenario: contamination_10_48 Running scenario: contamination_10_34 Running scenario: contamination_10_38 Running scenario: contamination_10_42 Running scenario: contamination_10_46 Running scenario: contamination_10_33 Running scenario: contamination_10_37 Running scenario: contamination_10_41 Running scenario: contamination_10_45 Running scenario: contamination_10_19 Running scenario: contamination_10_20 Running scenario: contamination_10_23 Running scenario: contamination_10_24 Running scenario: contamination_10_27 Running scenario: contamination_10_28 Running scenario: contamination_10_31 Running scenario: contamination_10_32 Running scenario: contamination_10_18 Running scenario: contamination_10_22 Running scenario: contamination_10_26 Running scenario: contamination_10_30 Running scenario: contamination_10_17 Running scenario: contamination_10_21 Running scenario: contamination_10_25 Running scenario: contamination_10_29 Running scenario: contamination_25_10 Running scenario: contamination_25_12 Running scenario: contamination_25_14 Running scenario: contamination_25_16 Running scenario: contamination_25_09 Running scenario: contamination_25_11 Running scenario: contamination_25_13 Running scenario: contamination_25_15 Running scenario: contamination_25_02 Running scenario: contamination_25_04 Running scenario: contamination_25_06 Running scenario: contamination_25_08 Running scenario: contamination_25_01 Running scenario: contamination_25_03 Running scenario: contamination_25_05 Running scenario: contamination_25_07 Running scenario: contamination_25_35 Running scenario: contamination_25_36 Running scenario: contamination_25_39 Running scenario: contamination_25_40 Running scenario: contamination_25_43 Running scenario: contamination_25_44 Running scenario: contamination_25_47 Running scenario: contamination_25_48 Running scenario: contamination_25_34 Running scenario: contamination_25_38 Running scenario: contamination_25_42 Running scenario: contamination_25_46 Running scenario: contamination_25_33 Running scenario: contamination_25_37 Running scenario: contamination_25_41 Running scenario: contamination_25_45 Running scenario: contamination_25_19 Running scenario: contamination_25_20 Running scenario: contamination_25_23 Running scenario: contamination_25_24 Running scenario: contamination_25_27 Running scenario: contamination_25_28 Running scenario: contamination_25_31 Running scenario: contamination_25_32 Running scenario: contamination_25_18 Running scenario: contamination_25_22 Running scenario: contamination_25_26 Running scenario: contamination_25_30 Running scenario: contamination_25_17 Running scenario: contamination_25_21 Running scenario: contamination_25_25 Running scenario: contamination_25_29 Running scenario: contamination_5_10 Running scenario: contamination_5_12 Running scenario: contamination_5_14 Running scenario: contamination_5_16 Running scenario: contamination_5_09 Running scenario: contamination_5_11 Running scenario: contamination_5_13 Running scenario: contamination_5_15 Running scenario: contamination_5_02 Running scenario: contamination_5_04 Running scenario: contamination_5_06 Running scenario: contamination_5_08 Running scenario: contamination_5_01 Running scenario: contamination_5_03 Running scenario: contamination_5_05 Running scenario: contamination_5_07 Running scenario: contamination_5_35 Running scenario: contamination_5_36 Running scenario: contamination_5_39 Running scenario: contamination_5_40 Running scenario: contamination_5_43 Running scenario: contamination_5_44 Running scenario: contamination_5_47 Running scenario: contamination_5_48 Running scenario: contamination_5_34 Running scenario: contamination_5_38 Running scenario: contamination_5_42 Running scenario: contamination_5_46 Running scenario: contamination_5_33 Running scenario: contamination_5_37 Running scenario: contamination_5_41 Running scenario: contamination_5_45 Running scenario: contamination_5_19 Running scenario: contamination_5_20 Running scenario: contamination_5_23 Running scenario: contamination_5_24 Running scenario: contamination_5_27 Running scenario: contamination_5_28 Running scenario: contamination_5_31 Running scenario: contamination_5_32 Running scenario: contamination_5_18 Running scenario: contamination_5_22 Running scenario: contamination_5_26 Running scenario: contamination_5_30 Running scenario: contamination_5_17 Running scenario: contamination_5_21 Running scenario: contamination_5_25 Running scenario: contamination_5_29 Running scenario: large_10 Running scenario: large_12 Running scenario: large_14 Running scenario: large_16 Running scenario: large_09 Running scenario: large_11 Running scenario: large_13 Running scenario: large_15 Running scenario: large_02 Running scenario: large_04 Running scenario: large_06 Running scenario: large_08 Running scenario: large_01 Running scenario: large_03 Running scenario: large_05 Running scenario: large_07 Running scenario: large_35 Running scenario: large_36 Running scenario: large_39 Running scenario: large_40 Running scenario: large_43 Running scenario: large_44 Running scenario: large_47 Running scenario: large_48 Running scenario: large_34 Running scenario: large_38 Running scenario: large_42 Running scenario: large_46 Running scenario: large_33 Running scenario: large_37 Running scenario: large_41 Running scenario: large_45 Running scenario: large_19 Running scenario: large_20 Running scenario: large_23 Running scenario: large_24 Running scenario: large_27 Running scenario: large_28 Running scenario: large_31 Running scenario: large_32 Running scenario: large_18 Running scenario: large_22 Running scenario: large_26 Running scenario: large_30 Running scenario: large_17 Running scenario: large_21 Running scenario: large_25 Running scenario: large_29 Running scenario: maritime_10 Running scenario: maritime_12 Running scenario: maritime_14 Running scenario: maritime_16 Running scenario: maritime_09 Running scenario: maritime_11 Running scenario: maritime_13 Running scenario: maritime_15 Running scenario: maritime_02 Running scenario: maritime_04 Running scenario: maritime_06 Running scenario: maritime_08 Running scenario: maritime_01 Running scenario: maritime_03 Running scenario: maritime_05 Running scenario: maritime_07 Running scenario: maritime_35 Running scenario: maritime_36 Running scenario: maritime_39 Running scenario: maritime_40 Running scenario: maritime_43 Running scenario: maritime_44 Running scenario: maritime_47 Running scenario: maritime_48 Running scenario: maritime_34 Running scenario: maritime_38 Running scenario: maritime_42 Running scenario: maritime_46 Running scenario: maritime_33 Running scenario: maritime_37 Running scenario: maritime_41 Running scenario: maritime_45 Running scenario: maritime_19 Running scenario: maritime_20 Running scenario: maritime_23 Running scenario: maritime_24 Running scenario: maritime_27 Running scenario: maritime_28 Running scenario: maritime_31 Running scenario: maritime_32 Running scenario: maritime_18 Running scenario: maritime_22 Running scenario: maritime_26 Running scenario: maritime_30 Running scenario: maritime_17 Running scenario: maritime_21 Running scenario: maritime_25 Running scenario: maritime_29 Running scenario: small_10 Running scenario: small_12 Running scenario: small_14 Running scenario: small_16 Running scenario: small_09 Running scenario: small_11 Running scenario: small_13 Running scenario: small_15 Running scenario: small_02 Running scenario: small_04 Running scenario: small_06 Running scenario: small_08 Running scenario: small_01 Running scenario: small_03 Running scenario: small_05 Running scenario: small_07 Running scenario: small_35 Running scenario: small_36 Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Warning: Within box proportion is too low and/or interval is too high to achieve sample size. Automatically increasing within box proportion to achieve sample size. Running scenario: small_39 Running scenario: small_40 Running scenario: small_43 Running scenario: small_44 Running scenario: small_47 Running scenario: small_48 Running scenario: small_34 Running scenario: small_38 Running scenario: small_42 Running scenario: small_46 Running scenario: small_33 Running scenario: small_37 Running scenario: small_41 Running scenario: small_45 Running scenario: small_19 Running scenario: small_20 Running scenario: small_23 Running scenario: small_24 Running scenario: small_27 Running scenario: small_28 Running scenario: small_31 Running scenario: small_32 Running scenario: small_18 Running scenario: small_22 Running scenario: small_26 Running scenario: small_30 Running scenario: small_17 Running scenario: small_21 Running scenario: small_25 Running scenario: small_29
In [5]:
df = save_scenario_result_to_pandas(
results,
config_columns=[
"name",
"consignment_name",
"inspection_name",
"consignment/boxes/min",
"consignment/boxes/max",
"consignment/items_per_box/default",
"contamination/contamination_rate/parameters",
"contamination/arrangement",
"inspection/unit",
"inspection/sample_strategy",
"inspection/proportion/value",
"inspection/hypergeometric/detection_level",
"inspection/selection_strategy",
"inspection/within_box_pct",
"inspection/cluster/cluster_selection",
],
result_columns=[
"missing",
"true_contamination_rate",
"max_missed_contamination_rate",
"avg_missed_contamination_rate",
"max_intercepted_contamination_rate",
"avg_intercepted_contamination_rate",
"avg_boxes_opened_completion",
"avg_boxes_opened_detection",
"avg_items_inspected_completion",
"avg_items_inspected_detection",
"pct_items_inspected_detection",
"pct_contaminant_unreported_if_detection",
],
)
In [6]:
df['consignment_name'] = pd.Categorical(df['consignment_name'])
consignment_order = ["contamination_5", "contamination_10", "contamination_25", "small", "large", "maritime", "clustered", "clustered_small", "clustered_large"]
df['consignment_name'] = df['consignment_name'].cat.reorder_categories(consignment_order, ordered=True)
df
Out[6]:
| name | consignment_name | inspection_name | consignment/boxes/min | consignment/boxes/max | consignment/items_per_box/default | contamination/contamination_rate/parameters | contamination/arrangement | inspection/unit | inspection/sample_strategy | ... | max_missed_contamination_rate | avg_missed_contamination_rate | max_intercepted_contamination_rate | avg_intercepted_contamination_rate | avg_boxes_opened_completion | avg_boxes_opened_detection | avg_items_inspected_completion | avg_items_inspected_detection | pct_items_inspected_detection | pct_contaminant_unreported_if_detection | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | clustered_10 | clustered | boxes 0.01 hypergeometric random | 5 | 25 | 200 | [1, 80] | clustered | box | hypergeometric | ... | NaN | NaN | 0.018177 | 0.007520 | 15.95 | 2.05 | 3190.0 | 410.0 | 12.959031 | 0.00000 |
| 1 | clustered_12 | clustered | boxes 0.05 hypergeometric random | 5 | 25 | 200 | [1, 80] | clustered | box | hypergeometric | ... | NaN | NaN | 0.018242 | 0.007540 | 14.55 | 1.80 | 2910.0 | 360.0 | 12.645243 | 0.00000 |
| 2 | clustered_14 | clustered | boxes 0.1 hypergeometric random | 5 | 25 | 200 | [1, 80] | clustered | box | hypergeometric | ... | 0.010500 | 0.010500 | 0.018186 | 0.007436 | 11.10 | 2.50 | 2220.0 | 500.0 | 19.696030 | 0.00000 |
| 3 | clustered_16 | clustered | boxes 0.25 hypergeometric random | 5 | 25 | 200 | [1, 80] | clustered | box | hypergeometric | ... | 0.013576 | 0.011662 | 0.017649 | 0.006498 | 7.55 | 3.30 | 1510.0 | 660.0 | 24.203297 | 0.00000 |
| 4 | clustered_09 | clustered | boxes 0.01 hypergeometric convenience | 5 | 25 | 200 | [1, 80] | clustered | box | hypergeometric | ... | NaN | NaN | 0.018095 | 0.007498 | 14.20 | 1.80 | 2840.0 | 360.0 | 12.176101 | 0.00000 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 427 | small_30 | small | items 25% random | 1 | 3 | 200 | [1, 80] | random | item | proportion | ... | 0.030000 | 0.010301 | NaN | NaN | 2.20 | 2.20 | 8.8 | 8.8 | 2.000000 | 0.00000 |
| 428 | small_17 | small | items 2% convenience | 1 | 3 | 200 | [1, 80] | random | item | proportion | ... | 0.025000 | 0.012396 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 429 | small_21 | small | items 5% convenience | 1 | 3 | 200 | [1, 80] | random | item | proportion | ... | 0.025000 | 0.012396 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 430 | small_25 | small | items 10% convenience | 1 | 3 | 200 | [1, 80] | random | item | proportion | ... | 0.025000 | 0.012396 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 431 | small_29 | small | items 25% convenience | 1 | 3 | 200 | [1, 80] | random | item | proportion | ... | 0.025000 | 0.012396 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
432 rows × 27 columns
In [7]:
df.plot(kind="bar", x="inspection_name", y="missing")
Out[7]:
<matplotlib.axes._subplots.AxesSubplot at 0x7f00bfa04160>
In [8]:
#df["consignment"] = df.apply(lambda row: re.sub("_[0-9]+$", "", row["name"]), axis = 1)
In [9]:
df.groupby("consignment_name").agg({"missing": "mean"}).reset_index().plot(kind="bar", x="consignment_name", y="missing")
Out[9]:
<matplotlib.axes._subplots.AxesSubplot at 0x7f00bd568790>
In [10]:
df.groupby("inspection_name").agg({"missing": "mean"}).reset_index().plot(kind="bar", x="inspection_name", y="missing")
plt.xlabel("inspection", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Avg percent contaminated consignments missed by inspection", size=16)
Out[10]:
Text(0.5, 1.0, 'Avg percent contaminated consignments missed by inspection')
In [11]:
inspections_vs_consignments = df.pivot_table(index=["inspection_name"], columns=["consignment_name"], values=["missing"])
inspections_vs_consignments
Out[11]:
| missing | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| consignment_name | contamination_5 | contamination_10 | contamination_25 | small | large | maritime | clustered | clustered_small | clustered_large |
| inspection_name | |||||||||
| boxes 0.01 hypergeometric convenience | 0.0 | 0.0 | 0.0 | 0.000000 | 0.0 | 0.000000 | 0.000000 | 0.000000 | 0.0 |
| boxes 0.01 hypergeometric random | 0.0 | 0.0 | 0.0 | 0.000000 | 10.0 | 0.000000 | 0.000000 | 0.000000 | 10.0 |
| boxes 0.05 hypergeometric convenience | 0.0 | 0.0 | 0.0 | 0.000000 | 40.0 | 5.555556 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.05 hypergeometric random | 5.0 | 0.0 | 0.0 | 0.000000 | 50.0 | 0.000000 | 0.000000 | 0.000000 | 35.0 |
| boxes 0.1 hypergeometric convenience | 5.0 | 0.0 | 0.0 | 0.000000 | 75.0 | 10.555556 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.1 hypergeometric random | 5.0 | 0.0 | 0.0 | 0.000000 | 50.0 | 5.000000 | 5.000000 | 0.000000 | 80.0 |
| boxes 0.25 hypergeometric convenience | 30.0 | 20.0 | 0.0 | 0.000000 | 90.0 | 36.666667 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.25 hypergeometric random | 45.0 | 25.0 | 0.0 | 0.000000 | 80.0 | 41.666667 | 30.555556 | 0.000000 | 100.0 |
| boxes 10% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 10% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 2% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 2% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 25% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 25% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 5% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 5% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| items 0.01 hypergeometric cluster convenience | 75.0 | 75.0 | 40.0 | 6.250000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.01 hypergeometric cluster random | 80.0 | 75.0 | 45.0 | 0.000000 | 100.0 | 100.000000 | 89.444444 | 6.250000 | 95.0 |
| items 0.01 hypergeometric convenience | 75.0 | 75.0 | 45.0 | 12.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.01 hypergeometric random | 0.0 | 0.0 | 0.0 | 24.285714 | 20.0 | 16.111111 | 32.222222 | 32.539683 | 25.0 |
| items 0.05 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.05 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.05 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.05 hypergeometric random | 10.0 | 0.0 | 0.0 | 76.388889 | 70.0 | 67.222222 | 53.333333 | 56.349206 | 60.0 |
| items 0.1 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.1 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.1 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.1 hypergeometric random | 25.0 | 10.0 | 0.0 | 61.904762 | 75.0 | 79.444444 | 73.888889 | 69.047619 | 80.0 |
| items 0.25 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.25 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.25 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.25 hypergeometric random | 50.0 | 30.0 | 5.0 | 94.444444 | 95.0 | 89.444444 | 95.000000 | 87.301587 | 85.0 |
| items 10% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 10% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 10% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 10% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 2% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 2% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 2% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 2% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 25% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 25% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 25% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 25% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 5% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 5% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 5% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 5% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
In [12]:
inspections_vs_consignments.plot(kind="bar")
Out[12]:
<matplotlib.axes._subplots.AxesSubplot at 0x7f00bd0d9190>
In [13]:
ax = inspections_vs_consignments.plot(kind="bar", figsize=(100, 10))
#ax.figure.savefig("inspections_vs_consignments_bar.png")
In [14]:
from itertools import cycle, islice
it = cycle(list(range(1,49)))
df['inspection_num'] = list(islice(it, len(df)))
In [15]:
plt.figure(figsize=(30, 8))
sns.boxplot(x="inspection_name", y="missing", data=df)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.xlabel("inspection", size=16)
plt.ylabel("percent missed", size=16)
plt.title("Avg. percent contaminated consignments missed by inspection", size=20)
plt.savefig("missed_by_inspection_boxplot.png", bbox_inches = "tight")
In [16]:
plt.figure(figsize=(30, 8))
sns.barplot(x="inspection_name", y="missing", data=df)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.xlabel("inspection", size=16)
plt.ylabel("percent missed", size=16)
plt.title("Avg. percent contaminated consignments missed by inspection", size=20)
plt.savefig("missed_by_inspection.png", bbox_inches = "tight")
In [17]:
sns.barplot(x="consignment_name", y="missing", data=df)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.xlabel("consignment type", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Avg. percent contaminated consignments missed by consignment type", size=16)
plt.savefig("missed_by_consignment.png", bbox_inches = "tight")
In [18]:
sns.boxplot(x="consignment_name", y="missing", data=df)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.xlabel("consignment type", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Percent contaminated consignments missed by consignment type", size=16)
plt.savefig("missed_by_inspection_boxplot.png", bbox_inches = "tight")
In [19]:
sns.catplot(x="consignment_name", y="missing", hue="inspection/sample_strategy", kind="box", data=df, height=8.27, aspect=11.7/8.27)
plt.xlabel("consignment type", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Percent contaminated consignments missed by consignment type and sample strategy", size=16)
plt.savefig("missed_by_consignment_sample_strategy.png", bbox_inches = "tight")
In [20]:
sns.catplot(x="consignment_name", y="missing", hue="inspection/unit", kind="box", data=df, height=9, aspect=11.7/8.27)
plt.xlabel("consignment type", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Percent contaminated consignments missed by consignment type and inspection unit", size=16)
plt.savefig("missed_by_consignment_unit.png", bbox_inches = "tight")
In [21]:
chart = sns.catplot(x="consignment_name", y="missing", hue="inspection/selection_strategy", aspect=11.7/8.27,
kind="box", data=df, height=8)
chart.set_xticklabels(rotation=65, horizontalalignment='right')
plt.xlabel("consignment type", size=12)
plt.ylabel("percent missed", size=12)
plt.title("Percent contaminated consignments missed by consignment type and selection strategy", size=16)
chart.savefig("missed_by_consignment_selection.png", bbox_inches = "tight")
In [22]:
inspections_vs_consignments
Out[22]:
| missing | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| consignment_name | contamination_5 | contamination_10 | contamination_25 | small | large | maritime | clustered | clustered_small | clustered_large |
| inspection_name | |||||||||
| boxes 0.01 hypergeometric convenience | 0.0 | 0.0 | 0.0 | 0.000000 | 0.0 | 0.000000 | 0.000000 | 0.000000 | 0.0 |
| boxes 0.01 hypergeometric random | 0.0 | 0.0 | 0.0 | 0.000000 | 10.0 | 0.000000 | 0.000000 | 0.000000 | 10.0 |
| boxes 0.05 hypergeometric convenience | 0.0 | 0.0 | 0.0 | 0.000000 | 40.0 | 5.555556 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.05 hypergeometric random | 5.0 | 0.0 | 0.0 | 0.000000 | 50.0 | 0.000000 | 0.000000 | 0.000000 | 35.0 |
| boxes 0.1 hypergeometric convenience | 5.0 | 0.0 | 0.0 | 0.000000 | 75.0 | 10.555556 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.1 hypergeometric random | 5.0 | 0.0 | 0.0 | 0.000000 | 50.0 | 5.000000 | 5.000000 | 0.000000 | 80.0 |
| boxes 0.25 hypergeometric convenience | 30.0 | 20.0 | 0.0 | 0.000000 | 90.0 | 36.666667 | 0.000000 | 0.000000 | 5.0 |
| boxes 0.25 hypergeometric random | 45.0 | 25.0 | 0.0 | 0.000000 | 80.0 | 41.666667 | 30.555556 | 0.000000 | 100.0 |
| boxes 10% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 10% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 2% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 2% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 25% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 25% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| boxes 5% convenience | 100.0 | 100.0 | 100.0 | 100.000000 | 90.0 | 100.000000 | 100.000000 | 100.000000 | 5.0 |
| boxes 5% random | 100.0 | 100.0 | 100.0 | 100.000000 | 100.0 | 100.000000 | 100.000000 | 100.000000 | 100.0 |
| items 0.01 hypergeometric cluster convenience | 75.0 | 75.0 | 40.0 | 6.250000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.01 hypergeometric cluster random | 80.0 | 75.0 | 45.0 | 0.000000 | 100.0 | 100.000000 | 89.444444 | 6.250000 | 95.0 |
| items 0.01 hypergeometric convenience | 75.0 | 75.0 | 45.0 | 12.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.01 hypergeometric random | 0.0 | 0.0 | 0.0 | 24.285714 | 20.0 | 16.111111 | 32.222222 | 32.539683 | 25.0 |
| items 0.05 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.05 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.05 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.05 hypergeometric random | 10.0 | 0.0 | 0.0 | 76.388889 | 70.0 | 67.222222 | 53.333333 | 56.349206 | 60.0 |
| items 0.1 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.1 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.1 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.1 hypergeometric random | 25.0 | 10.0 | 0.0 | 61.904762 | 75.0 | 79.444444 | 73.888889 | 69.047619 | 80.0 |
| items 0.25 hypergeometric cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.25 hypergeometric cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 0.25 hypergeometric convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 95.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 0.25 hypergeometric random | 50.0 | 30.0 | 5.0 | 94.444444 | 95.0 | 89.444444 | 95.000000 | 87.301587 | 85.0 |
| items 10% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 10% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 10% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 10% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 2% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 2% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 2% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 2% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 25% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 25% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 25% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 25% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
| items 5% cluster convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 5% cluster random | 85.0 | 80.0 | 65.0 | 31.250000 | 100.0 | 100.000000 | 94.444444 | 38.888889 | 100.0 |
| items 5% convenience | 85.0 | 85.0 | 60.0 | 37.500000 | 90.0 | 83.888889 | 0.000000 | 0.000000 | 5.0 |
| items 5% random | 10.0 | 5.0 | 0.0 | 100.000000 | 15.0 | 20.555556 | 63.888889 | 88.194444 | 10.0 |
In [23]:
plt.figure(figsize=(10, 10))
sns.heatmap(data=inspections_vs_consignments, annot=True, fmt=".2f", xticklabels = consignment_order)
plt.xlabel("consignment type", size=12)
plt.ylabel("inspection", size=12)
plt.title("Percent contaminated consignments missed", size=16)
plt.savefig("heatmap_missed.png", bbox_inches = "tight")
In [24]:
inspections_vs_consignments_boxes_opened = df.pivot_table(index=["inspection_name"], columns=["consignment_name"], values=["avg_boxes_opened_completion"])
plt.figure(figsize=(10, 10))
sns.heatmap(data=inspections_vs_consignments_boxes_opened, annot=True, fmt=".2f", xticklabels = consignment_order)
plt.xlabel("consignment type", size=12)
plt.ylabel("inspection", size=12)
plt.title("Avg. number of boxes opened (inspect to completion)", size=16)
plt.savefig("heatmap_boxes_completion.png", bbox_inches = "tight")
In [25]:
inspections_vs_consignments_items_inspected = df.pivot_table(index=["inspection_name"], columns=["consignment_name"], values=["avg_items_inspected_completion"])
plt.figure(figsize=(10, 10))
sns.heatmap(data=inspections_vs_consignments_items_inspected, annot=True, fmt=".2f", xticklabels = consignment_order)
plt.xlabel("consignment type", size=12)
plt.ylabel("inspection", size=12)
plt.title("Avg. number of items inpected (inspect to completion)", size=16)
plt.savefig("heatmap_items_completion.png", bbox_inches = "tight")
In [26]:
inspections_vs_consignments_boxes_opened_detection = df.pivot_table(index=["inspection_name"], columns=["consignment_name"], values=["avg_boxes_opened_detection"])
plt.figure(figsize=(10, 10))
sns.heatmap(data=inspections_vs_consignments_boxes_opened_detection, annot=True, fmt=".2f", xticklabels = consignment_order)
plt.xlabel("consignment type", size=12)
plt.ylabel("inspection", size=12)
plt.title("Avg. number of boxes opened (inspect to detection)", size=16)
plt.savefig("heatmap_boxes_detection.png", bbox_inches = "tight")
In [27]:
inspections_vs_consignments_items_inspected_detection = df.pivot_table(index=["inspection_name"], columns=["consignment_name"], values=["avg_items_inspected_detection"])
plt.figure(figsize=(10, 10))
sns.heatmap(data=inspections_vs_consignments_items_inspected_detection, annot=True, fmt=".2f", xticklabels = consignment_order)
plt.xlabel("consignment type", size=12)
plt.ylabel("inspection", size=12)
plt.title("Avg. number of items inspected (inspect to detection)", size=16)
plt.savefig("heatmap_items_detection.png", bbox_inches = "tight")
In [28]:
df_norm = df[["missing",
"true_contamination_rate",
"max_intercepted_contamination_rate",
"avg_intercepted_contamination_rate",
"avg_boxes_opened_completion",
"avg_boxes_opened_detection",
"avg_items_inspected_completion",
"avg_items_inspected_detection",
"pct_items_inspected_detection",
"pct_contaminant_unreported_if_detection"]]
df_norm
Out[28]:
| missing | true_contamination_rate | max_intercepted_contamination_rate | avg_intercepted_contamination_rate | avg_boxes_opened_completion | avg_boxes_opened_detection | avg_items_inspected_completion | avg_items_inspected_detection | pct_items_inspected_detection | pct_contaminant_unreported_if_detection | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.000000 | 0.007169 | 0.018177 | 0.007520 | 15.95 | 2.05 | 3190.0 | 410.0 | 12.959031 | 0.00000 |
| 1 | 0.000000 | 0.007184 | 0.018242 | 0.007540 | 14.55 | 1.80 | 2910.0 | 360.0 | 12.645243 | 0.00000 |
| 2 | 5.000000 | 0.007217 | 0.018186 | 0.007436 | 11.10 | 2.50 | 2220.0 | 500.0 | 19.696030 | 0.00000 |
| 3 | 30.555556 | 0.007112 | 0.017649 | 0.006498 | 7.55 | 3.30 | 1510.0 | 660.0 | 24.203297 | 0.00000 |
| 4 | 0.000000 | 0.007144 | 0.018095 | 0.007498 | 14.20 | 1.80 | 2840.0 | 360.0 | 12.176101 | 0.00000 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 427 | 100.000000 | 0.007417 | NaN | NaN | 2.20 | 2.20 | 8.8 | 8.8 | 2.000000 | 0.00000 |
| 428 | 37.500000 | 0.007833 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 429 | 37.500000 | 0.007833 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 430 | 37.500000 | 0.007833 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
| 431 | 37.500000 | 0.007833 | 0.016250 | 0.007917 | 1.00 | 1.00 | 8.2 | 5.1 | 1.236905 | 60.25641 |
432 rows × 10 columns
In [29]:
df_norm=(df_norm-df_norm.min())/(df_norm.max()-df_norm.min())
df_norm
Out[29]:
| missing | true_contamination_rate | max_intercepted_contamination_rate | avg_intercepted_contamination_rate | avg_boxes_opened_completion | avg_boxes_opened_detection | avg_items_inspected_completion | avg_items_inspected_detection | pct_items_inspected_detection | pct_contaminant_unreported_if_detection | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.000000 | 0.001488 | 0.047963 | 0.011717 | 0.058998 | 0.029624 | 0.095509 | 0.030920 | 0.166446 | 0.000000 |
| 1 | 0.000000 | 0.001545 | 0.048191 | 0.011795 | 0.053819 | 0.026012 | 0.087126 | 0.027149 | 0.162416 | 0.000000 |
| 2 | 0.050000 | 0.001677 | 0.047996 | 0.011388 | 0.041058 | 0.036127 | 0.066467 | 0.037707 | 0.252977 | 0.000000 |
| 3 | 0.305556 | 0.001262 | 0.046111 | 0.007733 | 0.027927 | 0.047688 | 0.045210 | 0.049774 | 0.310868 | 0.000000 |
| 4 | 0.000000 | 0.001390 | 0.047676 | 0.011629 | 0.052525 | 0.026012 | 0.085030 | 0.027149 | 0.156390 | 0.000000 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 427 | 1.000000 | 0.002462 | NaN | NaN | 0.008138 | 0.031792 | 0.000263 | 0.000664 | 0.025688 | 0.000000 |
| 428 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 |
| 429 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 |
| 430 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 |
| 431 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 |
432 rows × 10 columns
In [30]:
df_norm["name"], df_norm["inspection_num"], df_norm["inspection_name"], df_norm["consignment_name"] = df["name"], df["inspection_num"], df["inspection_name"], df["consignment_name"]
df_norm
Out[30]:
| missing | true_contamination_rate | max_intercepted_contamination_rate | avg_intercepted_contamination_rate | avg_boxes_opened_completion | avg_boxes_opened_detection | avg_items_inspected_completion | avg_items_inspected_detection | pct_items_inspected_detection | pct_contaminant_unreported_if_detection | name | inspection_num | inspection_name | consignment_name | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.000000 | 0.001488 | 0.047963 | 0.011717 | 0.058998 | 0.029624 | 0.095509 | 0.030920 | 0.166446 | 0.000000 | clustered_10 | 1 | boxes 0.01 hypergeometric random | clustered |
| 1 | 0.000000 | 0.001545 | 0.048191 | 0.011795 | 0.053819 | 0.026012 | 0.087126 | 0.027149 | 0.162416 | 0.000000 | clustered_12 | 2 | boxes 0.05 hypergeometric random | clustered |
| 2 | 0.050000 | 0.001677 | 0.047996 | 0.011388 | 0.041058 | 0.036127 | 0.066467 | 0.037707 | 0.252977 | 0.000000 | clustered_14 | 3 | boxes 0.1 hypergeometric random | clustered |
| 3 | 0.305556 | 0.001262 | 0.046111 | 0.007733 | 0.027927 | 0.047688 | 0.045210 | 0.049774 | 0.310868 | 0.000000 | clustered_16 | 4 | boxes 0.25 hypergeometric random | clustered |
| 4 | 0.000000 | 0.001390 | 0.047676 | 0.011629 | 0.052525 | 0.026012 | 0.085030 | 0.027149 | 0.156390 | 0.000000 | clustered_09 | 5 | boxes 0.01 hypergeometric convenience | clustered |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 427 | 1.000000 | 0.002462 | NaN | NaN | 0.008138 | 0.031792 | 0.000263 | 0.000664 | 0.025688 | 0.000000 | small_30 | 44 | items 25% random | small |
| 428 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 | small_17 | 45 | items 2% convenience | small |
| 429 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 | small_21 | 46 | items 5% convenience | small |
| 430 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 | small_25 | 47 | items 10% convenience | small |
| 431 | 0.375000 | 0.004104 | 0.041199 | 0.013262 | 0.003699 | 0.014451 | 0.000246 | 0.000385 | 0.015887 | 0.603997 | small_29 | 48 | items 25% convenience | small |
432 rows × 14 columns
In [31]:
df_norm_melted = pd.melt(df_norm, id_vars=["inspection_num", "inspection_name", "consignment_name"], value_vars=["missing", "avg_boxes_opened_completion", "avg_items_inspected_completion"])
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted, ax=ax1, ci = None)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (mean value across all consignment types)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_all.png", bbox_inches = "tight")
In [32]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("contamination_5")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (5% Contamination consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_contamination_5.png", bbox_inches = "tight")
In [33]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("contamination_10")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (10% Contamination consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_contamination_10.png", bbox_inches = "tight")
In [34]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("contamination_25")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (25% Contamination consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_contamination_25.png", bbox_inches = "tight")
In [35]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("small")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Small consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_small.png", bbox_inches = "tight")
In [36]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("large")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Large consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_large.png", bbox_inches = "tight")
In [37]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("maritime")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Maritime consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_maritime.png", bbox_inches = "tight")
In [38]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("clustered")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Clustered consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_clustered.png", bbox_inches = "tight")
In [39]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("clustered_small")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Clustered small consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_clustered_small.png", bbox_inches = "tight")
In [40]:
fig, ax1 = plt.subplots(figsize=(30, 6))
chart = sns.barplot(x="inspection_name", y="value", hue="variable",\
data=df_norm_melted[df_norm_melted.consignment_name.eq("clustered_large")], ax=ax1)
plt.xticks(
rotation=45,
horizontalalignment='right',
fontweight='light',
fontsize='large'
)
plt.title("Normalized metrics by inspection (Clustered large consignments)")
plt.xlabel("inspection description")
plt.ylabel("normalized value")
plt.savefig("metrics_by_inspection_clustered_large.png", bbox_inches = "tight")
