Driverless AI: Credit Card Demo

This notebook provides an H2OAI Client workflow, of model building and scoring, that parallels the Driverless AI workflow.

Here is the Python Client Documentation.

py-client

Workflow Steps

Build an Experiment with Python API:

  1. Sign in

  2. Import train & test set/new data

  3. Specify experiment parameters

  4. Launch Experiment

  5. Examine Experiment

  6. Download Predictions

Build an Experiment in Web UI and Access Through Python:

  1. Get pointer to experiment

Score on New Data:

  1. Score on new data with H2OAI model

Model Diagnostics on New Data:

  1. Run model diagnostics on new data with H2O AI model

Run Model Interpretation

  1. Run model interpretation on the raw features

  2. Run Model Interpretation on External Model Predictions

Build Scoring Pipelines

  1. Build Python Scoring Pipeline

  2. Build MOJO Scoring Pipeline

Build an Experiment with Python API

1. Sign In

Import the required modules and log in.

Pass in your credentials through the Client class which creates an authentication token to send to the Driverless AI Server. In plain English: to sign into the Driverless AI web page (which then sends requests to the Driverless Server), instantiate the Client class with your Driverless AI address and login credentials.

[2]:
import driverlessai
import matplotlib.pyplot as plt
import pandas as pd
[3]:
address = 'http://ip_where_driverless_is_running:12345'
username = 'username'
password = 'password'
dai = driverlessai.Client(address = address, username = username, password = password)
# make sure to use the same user name and password when signing in through the GUI

Equivalent Steps in Driverless: Signing In

Equivalent Steps in Driverless: Signing In1

Equivalent Steps in Driverless: Signing In2

2. Upload Datasets

Upload training and testing datasets from the Driverless AI /data folder.

You can provide a training, validation, and testing dataset for an experiment. The validation and testing dataset are optional. In this example, we will provide only training and testing.

[4]:
train_path = 's3://h2o-public-test-data/smalldata/creditcard/CreditCard_Cat-train.csv'
test_path = 's3://h2o-public-test-data/smalldata/creditcard/CreditCard_Cat-test.csv'

train = dai.datasets.create(data=train_path, data_source='s3')
test = dai.datasets.create(data=test_path, data_source='s3')
Complete 100.00% - [4/4] Computing column statistics
Complete 100.00% - [4/4] Computing column statistics

Equivalent Steps in Driverless: Uploading Train & Test CSV Files

Equivalent Steps in Driverless: Upload Train & Test CSV Files

3. Set Experiment Parameters

We will now set the parameters of our experiment. Some of the parameters include:

  • Target Column: The column we are trying to predict.

  • Dropped Columns: The columns we do not want to use as predictors such as ID columns, columns with data leakage, etc.

  • Weight Column: The column that indicates the per row observation weights. If None, each row will have an observation weight of 1.

  • Fold Column: The column that indicates the fold. If None, the folds will be determined by Driverless AI.

  • Is Time Series: Whether or not the experiment is a time-series use case.

For information on the experiment settings, refer to the Experiment Settings.

For this example, we will be predicting ``default payment next month``. The parameters that control the experiment process are: accuracy, time, and interpretability. We can use the get_experiment_preview_sync function to get a sense of what will happen during the experiment.

We will start out by seeing what the experiment will look like with accuracy, time, and interpretability all set to 5.

[5]:
target = "DEFAULT_PAYMENT_NEXT_MONTH"
exp_preview = dai.experiments.preview(train_dataset=train,
                                      task='classification',
                                      target_column=target,
                                      accuracy=5,
                                      time=5,
                                      interpretability=5,
                                      config_overrides=None)
exp_preview
ACCURACY [5/10]:
- Training data size: *23,999 rows, 25 cols*
- Feature evolution: *[Constant, LightGBM, XGBoostGBM]*, *1/4 validation split*
- Final pipeline: *Ensemble (8 models), 4-fold CV*

TIME [5/10]:
- Feature evolution: *4 individuals*, up to *66 iterations*
- Early stopping: After *10* iterations of no improvement

INTERPRETABILITY [5/10]:
- Feature pre-pruning strategy: None
- Monotonicity constraints: disabled
- Feature engineering search space: [CVCatNumEncode, CVTargetEncode, CatOriginal, Cat, ClusterDist, ClusterTE, Frequent, Interactions, NumCatTE, NumToCatTE, NumToCatWoE, Original, TextLinModel, Text, TruncSVDNum, WeightOfEvidence]

[Constant, LightGBM, XGBoostGBM] models to train:
- Model and feature tuning: *16*
- Feature evolution: *104*
- Final pipeline: *8*

Estimated runtime: *minutes*
Auto-click Finish/Abort if not done in: *1 day*/*7 days*

With these settings, the Driverless AI experiment will train about 124 models: * 16 for model and feature tuning * 104 for feature evolution * 8 for the final pipeline

When we start the experiment, we can either:

  • specify parameters

  • use Driverless AI to suggest parameters

Driverless AI can suggest the parameters based on the dataset and target column. Below we will use the ``get_experiment_tuning_suggestion`` to see what settings Driverless AI suggests.

Driverless AI has found that the best parameters are to set ``accuracy = 5``, ``time = 4``, ``interpretability = 6``. It has selected ``AUC`` as the scorer (this is the default scorer for binomial problems).

Equivalent Steps in Driverless: Set the Knobs, Configuration & Launch

Equivalent Steps in Driverless: Set the Knobs

4. Launch Experiment: Feature Engineering + Final Model Training

Launch the experiment using the parameters that Driverless AI suggested along with the testset, scorer, and seed that were added. We can launch the experiment with the suggested parameters or create our own.

[7]:
ex = dai.experiments.create(train_dataset=train,
                            test_dataset=test,
                            target_column=target,
                            task='classification',
                            accuracy=5,
                            time=4,
                            interpretability=6,
                            scorer="AUC")
Experiment launched at: http://localhost:12345/#experiment?key=b7aba0d4-e235-11ea-9088-0242ac110002
Complete 100.00% - Status: Complete

Equivalent Steps in Driverless: Launch Experiment

Equivalent Steps in Driverless: Launch Your Experiment1

Equivalent Steps in Driverless: Launch Your Experiment2

5. Examine Experiment

View the final model score for the validation datasets. When feature engineering is complete, an ensemble model can be built depending on the accuracy setting. The experiment object also contains the score on the validation and test data for this ensemble model. In this case, the validation score is the score on the training cross-validation predictions.

[11]:
metrics = ex.metrics()

print("Final model Score on Validation Data: " + str(round(metrics['val_score'], 3)))
Final model Score on Validation Data: 0.774

6. Download Results

Once an experiment is complete, we can see that the UI presents us options of downloading the:

  • predictions

    • on the (holdout) train data

    • on the test data

  • experiment summary - summary of the experiment including feature importance

We will show an example of downloading the test predictions below. Note that equivalent commands can also be run for downloading the train (holdout) predictions.

[12]:
ex.artifacts.download(only=['test_predictions'], dst_dir='', overwrite=True,)
Downloaded 'test_preds.csv'
[12]:
{'test_predictions': 'test_preds.csv'}
[13]:
test_preds = pd.read_csv("./test_preds.csv")
test_preds.head()
[13]:
DEFAULT_PAYMENT_NEXT_MONTH.0 DEFAULT_PAYMENT_NEXT_MONTH.1
0 0.299739 0.700261
1 0.862815 0.137185
2 0.958172 0.041828
3 0.599148 0.400852
4 0.882068 0.117932

Build an Experiment in Web UI and Access Through Python

It is also possible to use the Python API to examine an experiment that was started through the Web UI using the experiment key.

Experiments List

1. Get pointer to experiment

You can get a pointer to the experiment by referencing the experiment key in the Web UI.

[14]:
# Get list of experiments
experiment_list = dai.experiments.list()
experiment_list
[14]:
[<class 'driverlessai._experiments.Experiment'> b7aba0d4-e235-11ea-9088-0242ac110002 nivufeke,
 <class 'driverlessai._experiments.Experiment'> aed84080-e234-11ea-9088-0242ac110002 sewateru]
[12]:
# Get pointer to experiment
exp = experiment_list[0]
exp

Score on New Data

You can use the Python API to score on new data. This is equivalent to the SCORE ON ANOTHER DATASET button in the Web UI. The example below scores on the test data and then downloads the predictions.

Pass in any dataset that has the same columns as the original training set. If you passed a test set during the H2OAI model building step, the predictions already exist.

1. Score Using the H2OAI Model

The following shows the predicted probability of default for each record in the test.

[15]:
# Get unlabeled data to make predictions on
data = dai.datasets.create(data='s3://h2o-public-test-data/smalldata/creditcard/CreditCard_Cat-test.csv',
                           data_source='s3',
                           force=True)

target = "DEFAULT_PAYMENT_NEXT_MONTH"
prediction = ex.predict(data, [target])
pred_path = prediction.download('.')
pred_table = pd.read_csv(pred_path)
pred_table.head()
Complete 100.00% - [4/4] Computing column statistics
Complete
Downloaded './b7aba0d4-e235-11ea-9088-0242ac110002_preds_01963cda.csv'
[15]:
DEFAULT_PAYMENT_NEXT_MONTH.0 DEFAULT_PAYMENT_NEXT_MONTH.1
0 0.299739 0.700261
1 0.862815 0.137185
2 0.958172 0.041828
3 0.599148 0.400852
4 0.882068 0.117932

We can also get the contribution each feature had to the final prediction by setting pred_contribs = True. This will give us an idea of how each feature effects the predictions.

[16]:
prediction = ex.predict(data, [target], include_shap_values=True)

pred_contributions_path = prediction.download('.')
pred_contributions_table = pd.read_csv(pred_contributions_path)
pred_contributions_table.head()
Complete
Downloaded 'b7aba0d4-e235-11ea-9088-0242ac110002_preds_e92e68d4.csv'
[16]:
DEFAULT_PAYMENT_NEXT_MONTH.0 DEFAULT_PAYMENT_NEXT_MONTH.1 contrib_0_AGE contrib_10_PAY_3 contrib_11_PAY_4 contrib_12_PAY_5 contrib_13_PAY_6 contrib_14_PAY_AMT1 contrib_15_PAY_AMT2 contrib_16_PAY_AMT3 ... contrib_45_CVTE:PAY_1:PAY_2:PAY_4.0 contrib_46_WoE:AGE:LIMIT_BAL:PAY_1:PAY_2:PAY_3:PAY_5.0 contrib_47_WoE:PAY_3:PAY_5:PAY_6.0 contrib_48_ClusterTE:ClusterID20:PAY_1:PAY_4:PAY_5:PAY_AMT1:PAY_AMT6.0 contrib_4_BILL_AMT4 contrib_5_BILL_AMT5 contrib_6_BILL_AMT6 contrib_7_LIMIT_BAL contrib_8_PAY_1 contrib_bias
0 0.299739 0.700261 -0.051685 0.017082 -0.012476 -0.023556 -0.007033 0.016873 0.062995 -0.037905 ... 1.314687 0.019138 -0.039752 -0.004087 0.043153 -0.014567 -0.024331 0.024522 0.526365 -1.34867
1 0.862815 0.137185 0.044812 0.017206 0.001976 0.002324 -0.001514 0.030999 -0.087150 -0.092674 ... -0.437666 -0.079960 -0.119075 0.008777 0.052826 0.019449 0.061261 0.138121 -0.015459 -1.34867
2 0.958172 0.041828 -0.054909 0.003906 -0.003241 0.002495 -0.003469 0.087790 -0.315745 -0.047180 ... -0.455067 0.007856 -0.103681 0.036041 -0.032951 0.021452 -0.060518 -0.161278 -0.015956 -1.34867
3 0.599148 0.400852 -0.003953 0.016077 0.002315 0.040801 0.007007 0.024191 -0.020707 0.147297 ... 1.342796 0.066763 0.025450 0.003859 -0.020837 -0.050288 0.019743 -0.096786 0.371450 -1.34867
4 0.882068 0.117932 0.000450 -0.038284 -0.013007 -0.001538 -0.003652 0.036131 0.219968 0.124056 ... -0.519380 -0.006065 -0.109650 -0.029302 -0.116205 0.007022 0.004275 0.031202 -0.016736 -1.34867

5 rows × 53 columns

We will examine the contributions for our first record more closely.

[17]:
contrib = pd.DataFrame(pred_contributions_table.iloc[0][1:])
contrib.columns = ["contribution"]
contrib["abs_contribution"] = contrib.contribution.abs()
contrib.sort_values(by="abs_contribution", ascending=False)[["contribution"]].head()
[17]:
contribution
contrib_bias -1.348670
contrib_45_CVTE:PAY_1:PAY_2:PAY_4.0 1.314687
DEFAULT_PAYMENT_NEXT_MONTH.1 0.700261
contrib_8_PAY_1 0.526365
contrib_21_CVTE:EDUCATION.0 0.140235

The clusters from this customer’s: PAY_0, PAY_2, and LIMIT_BAL had the greatest impact on their prediction. Since the contribution is positive, we know that it increases the probability that they will default.

Build Scoring Pipelines

In our last section, we will build the scoring pipelines from our experiment. There are two scoring pipeline options:

  • Python Scoring Pipeline: requires Python runtime

  • MOJO Scoring Pipeline: requires Java runtime

Documentation on the scoring pipelines is provided here: http://docs.h2o.ai/driverless-ai/latest-stable/docs/userguide/python-mojo-pipelines.html.

Equivalent Steps in Driverless: View Results3

The experiment screen shows two scoring pipeline buttons: Download Python Scoring Pipeline or Build MOJO Scoring Pipeline. Driverless AI determines if any scoring pipeline should be automatically built based on the config.toml file. In this example, we have run Driverless AI with the settings:

# Whether to create the Python scoring pipeline at the end of each experiment
make_python_scoring_pipeline = true

# Whether to create the MOJO scoring pipeline at the end of each experiment
# Note: Not all transformers or main models are available for MOJO (e.g. no gblinear main model)
make_mojo_scoring_pipeline = false

Therefore, only the Python Scoring Pipeline will be built by default.

1. Build Python Scoring Pipeline

The Python Scoring Pipeline has been built by default based on our config.toml settings, but we can built it again if we wish - this is useful if the ``make_python_scoring_pipeline`` option was set to false.

[18]:
ex.artifacts.create('python_pipeline')
Building Python scoring pipeline...

Now we will download the scoring pipeline zip file.

[19]:
ex.artifacts.download(only='python_pipeline',
                      dst_dir='.',
                      overwrite=True)
Downloaded './scorer.zip'
[19]:
{'python_pipeline': './scorer.zip'}

2. Build MOJO Scoring Pipeline

The MOJO Scoring Pipeline has not been built by default because of our config.toml settings. We can build the MOJO Scoring Pipeline using the Python client. This is equivalent to selecting the Build MOJO Scoring Pipeline on the experiment screen.

[61]:
ex.artifacts.create('mojo_pipeline')

Now we can download the scoring pipeline zip file.

[63]:
ex.artifacts.download(only='mojo_pipeline', dst_dir='.', overwrite=True)
[63]:
'./mojo.zip'

Once the MOJO Scoring Pipeline is built, the Build MOJO Scoring Pipeline changes to Download MOJO Scoring Pipeline.

Equivalent Steps in Driverless: Download MOJO

[ ]: