Train sizing¶
This example shows how to asses a model's performance based on the size of the training set.
The data used is a variation on the Australian weather dataset from Kaggle. You can download it from here. The goal of this dataset is to predict whether or not it will rain tomorrow training a binary classifier on target RainTomorrow.
Load the data¶
In [1]:
Copied!
# Import packages
import pandas as pd
from atom import ATOMClassifier
# Import packages
import pandas as pd
from atom import ATOMClassifier
In [2]:
Copied!
# Load the data
X = pd.read_csv("./datasets/weatherAUS.csv")
# Let's have a look
X.head()
# Load the data
X = pd.read_csv("./datasets/weatherAUS.csv")
# Let's have a look
X.head()
Out[2]:
| Location | MinTemp | MaxTemp | Rainfall | Evaporation | Sunshine | WindGustDir | WindGustSpeed | WindDir9am | WindDir3pm | ... | Humidity9am | Humidity3pm | Pressure9am | Pressure3pm | Cloud9am | Cloud3pm | Temp9am | Temp3pm | RainToday | RainTomorrow | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | MelbourneAirport | 18.0 | 26.9 | 21.4 | 7.0 | 8.9 | SSE | 41.0 | W | SSE | ... | 95.0 | 54.0 | 1019.5 | 1017.0 | 8.0 | 5.0 | 18.5 | 26.0 | Yes | 0 |
| 1 | Adelaide | 17.2 | 23.4 | 0.0 | NaN | NaN | S | 41.0 | S | WSW | ... | 59.0 | 36.0 | 1015.7 | 1015.7 | NaN | NaN | 17.7 | 21.9 | No | 0 |
| 2 | Cairns | 18.6 | 24.6 | 7.4 | 3.0 | 6.1 | SSE | 54.0 | SSE | SE | ... | 78.0 | 57.0 | 1018.7 | 1016.6 | 3.0 | 3.0 | 20.8 | 24.1 | Yes | 0 |
| 3 | Portland | 13.6 | 16.8 | 4.2 | 1.2 | 0.0 | ESE | 39.0 | ESE | ESE | ... | 76.0 | 74.0 | 1021.4 | 1020.5 | 7.0 | 8.0 | 15.6 | 16.0 | Yes | 1 |
| 4 | Walpole | 16.4 | 19.9 | 0.0 | NaN | NaN | SE | 44.0 | SE | SE | ... | 78.0 | 70.0 | 1019.4 | 1018.9 | NaN | NaN | 17.4 | 18.1 | No | 0 |
5 rows × 22 columns
Run the pipeline¶
In [3]:
Copied!
# Initialize atom and prepare the data
atom = ATOMClassifier(X, verbose=2, warnings=False, random_state=1)
atom.clean()
atom.impute(strat_num="median", strat_cat="most_frequent", max_nan_rows=0.8)
atom.encode()
# Initialize atom and prepare the data
atom = ATOMClassifier(X, verbose=2, warnings=False, random_state=1)
atom.clean()
atom.impute(strat_num="median", strat_cat="most_frequent", max_nan_rows=0.8)
atom.encode()
<< ================== ATOM ================== >> Algorithm task: binary classification. Dataset stats ==================== >> Shape: (142193, 22) Memory: 61.69 MB Scaled: False Missing values: 316559 (10.1%) Categorical features: 5 (23.8%) Duplicate samples: 45 (0.0%) ------------------------------------- Train set size: 113755 Test set size: 28438 ------------------------------------- | | dataset | train | test | | - | -------------- | -------------- | -------------- | | 0 | 110316 (3.5) | 88253 (3.5) | 22063 (3.5) | | 1 | 31877 (1.0) | 25502 (1.0) | 6375 (1.0) | Cleaning the data... Fitting Imputer... Imputing missing values... --> Dropping 15182 samples for containing more than 16 missing values. --> Imputing 100 missing values with median (12.2) in feature MinTemp. --> Imputing 57 missing values with median (22.8) in feature MaxTemp. --> Imputing 640 missing values with median (0.0) in feature Rainfall. --> Imputing 46535 missing values with median (4.8) in feature Evaporation. --> Imputing 53034 missing values with median (8.5) in feature Sunshine. --> Imputing 4381 missing values with most_frequent (W) in feature WindGustDir. --> Imputing 4359 missing values with median (39.0) in feature WindGustSpeed. --> Imputing 6624 missing values with most_frequent (N) in feature WindDir9am. --> Imputing 612 missing values with most_frequent (SE) in feature WindDir3pm. --> Imputing 80 missing values with median (13.0) in feature WindSpeed9am. --> Imputing 49 missing values with median (19.0) in feature WindSpeed3pm. --> Imputing 532 missing values with median (69.0) in feature Humidity9am. --> Imputing 1168 missing values with median (52.0) in feature Humidity3pm. --> Imputing 1028 missing values with median (1017.6) in feature Pressure9am. --> Imputing 972 missing values with median (1015.2) in feature Pressure3pm. --> Imputing 42172 missing values with median (5.0) in feature Cloud9am. --> Imputing 44251 missing values with median (5.0) in feature Cloud3pm. --> Imputing 98 missing values with median (16.8) in feature Temp9am. --> Imputing 702 missing values with median (21.3) in feature Temp3pm. --> Imputing 640 missing values with most_frequent (No) in feature RainToday. Fitting Encoder... Encoding categorical columns... --> LeaveOneOut-encoding feature Location. Contains 45 classes. --> LeaveOneOut-encoding feature WindGustDir. Contains 16 classes. --> LeaveOneOut-encoding feature WindDir9am. Contains 16 classes. --> LeaveOneOut-encoding feature WindDir3pm. Contains 16 classes. --> Ordinal-encoding feature RainToday. Contains 2 classes.
In [4]:
Copied!
# Analyze the impact of the training set's size on a LightGBM model
atom.train_sizing("LGB", train_sizes=10, n_bootstrap=5)
# Analyze the impact of the training set's size on a LightGBM model
atom.train_sizing("LGB", train_sizes=10, n_bootstrap=5)
Run: 0 ================================ >> Size of training set: 10146 (10%) Size of test set: 25548 Training ========================= >> Models: LGB01 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.8195 Test evaluation --> f1: 0.6268 Time elapsed: 0.736s Bootstrap --------------------------------------- Evaluation --> f1: 0.6186 ± 0.0048 Time elapsed: 1.999s ------------------------------------------------- Total time: 2.736s Final results ==================== >> Duration: 2.736s ------------------------------------- LightGBM --> f1: 0.6186 ± 0.0048 ~ Run: 1 ================================ >> Size of training set: 20292 (20%) Size of test set: 25548 Training ========================= >> Models: LGB02 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.7403 Test evaluation --> f1: 0.6378 Time elapsed: 0.938s Bootstrap --------------------------------------- Evaluation --> f1: 0.6307 ± 0.002 Time elapsed: 2.556s ------------------------------------------------- Total time: 3.495s Final results ==================== >> Duration: 3.495s ------------------------------------- LightGBM --> f1: 0.6307 ± 0.002 Run: 2 ================================ >> Size of training set: 30438 (30%) Size of test set: 25548 Training ========================= >> Models: LGB03 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.7119 Test evaluation --> f1: 0.6452 Time elapsed: 1.134s Bootstrap --------------------------------------- Evaluation --> f1: 0.6342 ± 0.0017 Time elapsed: 3.172s ------------------------------------------------- Total time: 4.307s Final results ==================== >> Duration: 4.307s ------------------------------------- LightGBM --> f1: 0.6342 ± 0.0017 Run: 3 ================================ >> Size of training set: 40585 (40%) Size of test set: 25548 Training ========================= >> Models: LGB04 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6949 Test evaluation --> f1: 0.6451 Time elapsed: 1.407s Bootstrap --------------------------------------- Evaluation --> f1: 0.6372 ± 0.005 Time elapsed: 3.786s ------------------------------------------------- Total time: 5.194s Final results ==================== >> Duration: 5.195s ------------------------------------- LightGBM --> f1: 0.6372 ± 0.005 Run: 4 ================================ >> Size of training set: 50731 (50%) Size of test set: 25548 Training ========================= >> Models: LGB05 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6844 Test evaluation --> f1: 0.6459 Time elapsed: 1.608s Bootstrap --------------------------------------- Evaluation --> f1: 0.64 ± 0.0009 Time elapsed: 4.454s ------------------------------------------------- Total time: 6.066s Final results ==================== >> Duration: 6.067s ------------------------------------- LightGBM --> f1: 0.64 ± 0.0009 Run: 5 ================================ >> Size of training set: 60877 (60%) Size of test set: 25548 Training ========================= >> Models: LGB06 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6743 Test evaluation --> f1: 0.648 Time elapsed: 1.882s Bootstrap --------------------------------------- Evaluation --> f1: 0.6394 ± 0.0017 Time elapsed: 5.087s ------------------------------------------------- Total time: 6.971s Final results ==================== >> Duration: 6.972s ------------------------------------- LightGBM --> f1: 0.6394 ± 0.0017 Run: 6 ================================ >> Size of training set: 71024 (70%) Size of test set: 25548 Training ========================= >> Models: LGB07 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6723 Test evaluation --> f1: 0.6456 Time elapsed: 2.072s Bootstrap --------------------------------------- Evaluation --> f1: 0.6413 ± 0.0031 Time elapsed: 5.842s ------------------------------------------------- Total time: 7.916s Final results ==================== >> Duration: 7.917s ------------------------------------- LightGBM --> f1: 0.6413 ± 0.0031 Run: 7 ================================ >> Size of training set: 81170 (80%) Size of test set: 25548 Training ========================= >> Models: LGB08 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6677 Test evaluation --> f1: 0.6466 Time elapsed: 2.303s Bootstrap --------------------------------------- Evaluation --> f1: 0.6435 ± 0.0005 Time elapsed: 6.442s ------------------------------------------------- Total time: 8.747s Final results ==================== >> Duration: 8.747s ------------------------------------- LightGBM --> f1: 0.6435 ± 0.0005 Run: 8 ================================ >> Size of training set: 91316 (90%) Size of test set: 25548 Training ========================= >> Models: LGB09 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6669 Test evaluation --> f1: 0.6462 Time elapsed: 2.530s Bootstrap --------------------------------------- Evaluation --> f1: 0.6427 ± 0.0012 Time elapsed: 6.926s ------------------------------------------------- Total time: 9.458s Final results ==================== >> Duration: 9.458s ------------------------------------- LightGBM --> f1: 0.6427 ± 0.0012 Run: 9 ================================ >> Size of training set: 101463 (100%) Size of test set: 25548 Training ========================= >> Models: LGB10 Metric: f1 Results for LightGBM: Fit --------------------------------------------- Train evaluation --> f1: 0.6658 Test evaluation --> f1: 0.6466 Time elapsed: 2.811s Bootstrap --------------------------------------- Evaluation --> f1: 0.6444 ± 0.0042 Time elapsed: 7.606s ------------------------------------------------- Total time: 10.421s Final results ==================== >> Duration: 10.421s ------------------------------------- LightGBM --> f1: 0.6444 ± 0.0042
Analyze the results¶
In [5]:
Copied!
# The results are now multi-index, where frac is the fraction
# of the training set used to fit the model. The model names
# end with the fraction as well (without the dot)
atom.results
# The results are now multi-index, where frac is the fraction
# of the training set used to fit the model. The model names
# end with the fraction as well (without the dot)
atom.results
Out[5]:
| metric_train | metric_test | time_fit | mean_bootstrap | std_bootstrap | time_bootstrap | time | ||
|---|---|---|---|---|---|---|---|---|
| frac | model | |||||||
| 0.1 | LGB01 | 0.665781 | 0.646603 | 0.736s | 0.618559 | 0.004824 | 1.999s | 2.736s |
| 0.2 | LGB02 | 0.665781 | 0.646603 | 0.938s | 0.630685 | 0.002021 | 2.556s | 3.495s |
| 0.3 | LGB03 | 0.665781 | 0.646603 | 1.134s | 0.634181 | 0.001720 | 3.172s | 4.307s |
| 0.4 | LGB04 | 0.665781 | 0.646603 | 1.407s | 0.637212 | 0.004984 | 3.786s | 5.194s |
| 0.5 | LGB05 | 0.665781 | 0.646603 | 1.608s | 0.639984 | 0.000927 | 4.454s | 6.066s |
| 0.6 | LGB06 | 0.665781 | 0.646603 | 1.882s | 0.639425 | 0.001749 | 5.087s | 6.971s |
| 0.7 | LGB07 | 0.665781 | 0.646603 | 2.072s | 0.641320 | 0.003079 | 5.842s | 7.916s |
| 0.8 | LGB08 | 0.665781 | 0.646603 | 2.303s | 0.643463 | 0.000544 | 6.442s | 8.747s |
| 0.9 | LGB09 | 0.665781 | 0.646603 | 2.530s | 0.642688 | 0.001249 | 6.926s | 9.458s |
| 1.0 | LGB10 | 0.665781 | 0.646603 | 2.811s | 0.644359 | 0.004202 | 7.606s | 10.421s |
In [6]:
Copied!
# Every model can be accessed through its name
atom.lgb05.waterfall_plot(show=6)
# Every model can be accessed through its name
atom.lgb05.waterfall_plot(show=6)
In [7]:
Copied!
# Plot the train sizing's results
atom.plot_learning_curve()
# Plot the train sizing's results
atom.plot_learning_curve()
