Predicting Columns in a Table - Deployment Optimization

This tutorial will cover how to perform the end-to-end AutoML process to create an optimized and deployable AutoGluon artifact for production usage.

This tutorial assumes you have already read Predicting Columns in a Table - Quick Start and Predicting Columns in a Table - In Depth.

Fitting a TabularPredictor

We will again use the AdultIncome dataset as in the previous tutorials and train a predictor to predict whether the person’s income exceeds $50,000 or not, which is recorded in the class column of this table.

from autogluon.tabular import TabularDataset, TabularPredictor
train_data = TabularDataset('https://autogluon.s3.amazonaws.com/datasets/Inc/train.csv')
label = 'class'
subsample_size = 500  # subsample subset of data for faster demo, try setting this to much larger values
train_data = train_data.sample(n=subsample_size, random_state=0)
train_data.head()
age workclass fnlwgt education education-num marital-status occupation relationship race sex capital-gain capital-loss hours-per-week native-country class
6118 51 Private 39264 Some-college 10 Married-civ-spouse Exec-managerial Wife White Female 0 0 40 United-States >50K
23204 58 Private 51662 10th 6 Married-civ-spouse Other-service Wife White Female 0 0 8 United-States <=50K
29590 40 Private 326310 Some-college 10 Married-civ-spouse Craft-repair Husband White Male 0 0 44 United-States <=50K
18116 37 Private 222450 HS-grad 9 Never-married Sales Not-in-family White Male 0 2339 40 El-Salvador <=50K
33964 62 Private 109190 Bachelors 13 Married-civ-spouse Exec-managerial Husband White Male 15024 0 40 United-States >50K 
save_path = 'agModels-predictClass-deployment'  # specifies folder to store trained models
predictor = TabularPredictor(label=label, path=save_path).fit(train_data)

Beginning AutoGluon training ...
AutoGluon will save models to "agModels-predictClass-deployment/"
AutoGluon Version:  0.6.1b20221213
Python Version:     3.8.10
Operating System:   Linux
Platform Machine:   x86_64
Platform Version:   #1 SMP Tue Nov 30 00:17:50 UTC 2021
Train Data Rows:    500
Train Data Columns: 14
Label Column: class
Preprocessing data ...
AutoGluon infers your prediction problem is: 'binary' (because only two unique label-values observed).
    2 unique label values:  [' >50K', ' <=50K']
    If 'binary' is not the correct problem_type, please manually specify the problem_type parameter during predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression'])
Selected class <--> label mapping:  class 1 =  >50K, class 0 =  <=50K
    Note: For your binary classification, AutoGluon arbitrarily selected which label-value represents positive ( >50K) vs negative ( <=50K) class.
    To explicitly set the positive_class, either rename classes to 1 and 0, or specify positive_class in Predictor init.
Using Feature Generators to preprocess the data ...
Fitting AutoMLPipelineFeatureGenerator...
    Available Memory:                    31599.62 MB
    Train Data (Original)  Memory Usage: 0.29 MB (0.0% of available memory)
    Inferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.
    Stage 1 Generators:
            Fitting AsTypeFeatureGenerator...
                    Note: Converting 1 features to boolean dtype as they only contain 2 unique values.
    Stage 2 Generators:
            Fitting FillNaFeatureGenerator...
    Stage 3 Generators:
            Fitting IdentityFeatureGenerator...
            Fitting CategoryFeatureGenerator...
                    Fitting CategoryMemoryMinimizeFeatureGenerator...
    Stage 4 Generators:
            Fitting DropUniqueFeatureGenerator...
    Types of features in original data (raw dtype, special dtypes):
            ('int', [])    : 6 | ['age', 'fnlwgt', 'education-num', 'capital-gain', 'capital-loss', ...]
            ('object', []) : 8 | ['workclass', 'education', 'marital-status', 'occupation', 'relationship', ...]
    Types of features in processed data (raw dtype, special dtypes):
            ('category', [])  : 7 | ['workclass', 'education', 'marital-status', 'occupation', 'relationship', ...]
            ('int', [])       : 6 | ['age', 'fnlwgt', 'education-num', 'capital-gain', 'capital-loss', ...]
            ('int', ['bool']) : 1 | ['sex']
    0.1s = Fit runtime
    14 features in original data used to generate 14 features in processed data.
    Train Data (Processed) Memory Usage: 0.03 MB (0.0% of available memory)
Data preprocessing and feature engineering runtime = 0.09s ...
AutoGluon will gauge predictive performance using evaluation metric: 'accuracy'
    To change this, specify the eval_metric parameter of Predictor()
Automatically generating train/validation split with holdout_frac=0.2, Train Rows: 400, Val Rows: 100
Fitting 13 L1 models ...
Fitting model: KNeighborsUnif ...
    0.73     = Validation score   (accuracy)
    0.61s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: KNeighborsDist ...
    0.65     = Validation score   (accuracy)
    0.6s     = Training   runtime
    0.01s    = Validation runtime
Fitting model: LightGBMXT ...
    0.83     = Validation score   (accuracy)
    1.25s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: LightGBM ...
    0.85     = Validation score   (accuracy)
    0.82s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: RandomForestGini ...
    0.84     = Validation score   (accuracy)
    1.08s    = Training   runtime
    0.06s    = Validation runtime
Fitting model: RandomForestEntr ...
    0.83     = Validation score   (accuracy)
    1.06s    = Training   runtime
    0.06s    = Validation runtime
Fitting model: CatBoost ...
    0.85     = Validation score   (accuracy)
    1.4s     = Training   runtime
    0.01s    = Validation runtime
Fitting model: ExtraTreesGini ...
    0.82     = Validation score   (accuracy)
    1.07s    = Training   runtime
    0.06s    = Validation runtime
Fitting model: ExtraTreesEntr ...
    0.81     = Validation score   (accuracy)
    1.06s    = Training   runtime
    0.06s    = Validation runtime
Fitting model: NeuralNetFastAI ...
    0.82     = Validation score   (accuracy)
    2.61s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: XGBoost ...
    0.87     = Validation score   (accuracy)
    0.26s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: NeuralNetTorch ...
    0.83     = Validation score   (accuracy)
    1.02s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: LightGBMLarge ...
    0.83     = Validation score   (accuracy)
    0.54s    = Training   runtime
    0.01s    = Validation runtime
Fitting model: WeightedEnsemble_L2 ...
    0.87     = Validation score   (accuracy)
    0.32s    = Training   runtime
    0.0s     = Validation runtime
AutoGluon training complete, total runtime = 14.27s ... Best model: "WeightedEnsemble_L2"
TabularPredictor saved. To load, use: predictor = TabularPredictor.load("agModels-predictClass-deployment/")

Next, load separate test data to demonstrate how to make predictions on new examples at inference time:

test_data = TabularDataset('https://autogluon.s3.amazonaws.com/datasets/Inc/test.csv')
y_test = test_data[label]  # values to predict
test_data.head()

Loaded data from: https://autogluon.s3.amazonaws.com/datasets/Inc/test.csv | Columns = 15 / 15 | Rows = 9769 -> 9769
age workclass fnlwgt education education-num marital-status occupation relationship race sex capital-gain capital-loss hours-per-week native-country class
0 31 Private 169085 11th 7 Married-civ-spouse Sales Wife White Female 0 0 20 United-States <=50K
1 17 Self-emp-not-inc 226203 12th 8 Never-married Sales Own-child White Male 0 0 45 United-States <=50K
2 47 Private 54260 Assoc-voc 11 Married-civ-spouse Exec-managerial Husband White Male 0 1887 60 United-States >50K
3 21 Private 176262 Some-college 10 Never-married Exec-managerial Own-child White Female 0 0 30 United-States <=50K
4 17 Private 241185 12th 8 Never-married Prof-specialty Own-child White Male 0 0 20 United-States <=50K

We use our trained models to make predictions on the new data:

predictor = TabularPredictor.load(save_path)  # unnecessary, just demonstrates how to load previously-trained predictor from file

y_pred = predictor.predict(test_data)
y_pred

0        <=50K
1        <=50K
2        <=50K
3        <=50K
4        <=50K
         ...
9764     <=50K
9765     <=50K
9766     <=50K
9767     <=50K
9768     <=50K
Name: class, Length: 9769, dtype: object

We can use leaderboard to evaluate the performance of each individual trained model on our labeled test data:

predictor.leaderboard(test_data, silent=True)
model score_test score_val pred_time_test pred_time_val fit_time pred_time_test_marginal pred_time_val_marginal fit_time_marginal stack_level can_infer fit_order
0 RandomForestGini 0.842870 0.84 0.137671 0.055894 1.080855 0.137671 0.055894 1.080855 1 True 5
1 CatBoost 0.842461 0.85 0.012603 0.005573 1.403570 0.012603 0.005573 1.403570 1 True 7
2 RandomForestEntr 0.841130 0.83 0.140647 0.060857 1.060027 0.140647 0.060857 1.060027 1 True 6
3 LightGBM 0.839799 0.85 0.014990 0.008039 0.824368 0.014990 0.008039 0.824368 1 True 4
4 XGBoost 0.837445 0.87 0.050143 0.007187 0.261149 0.050143 0.007187 0.261149 1 True 11
5 WeightedEnsemble_L2 0.837445 0.87 0.052607 0.007834 0.583509 0.002464 0.000648 0.322360 2 True 14
6 LightGBMXT 0.836421 0.83 0.010455 0.005912 1.248788 0.010455 0.005912 1.248788 1 True 3
7 ExtraTreesGini 0.834579 0.82 0.139147 0.060351 1.065567 0.139147 0.060351 1.065567 1 True 8
8 NeuralNetTorch 0.833555 0.83 0.056062 0.013697 1.024997 0.056062 0.013697 1.024997 1 True 12
9 ExtraTreesEntr 0.833350 0.81 0.140015 0.058261 1.058253 0.140015 0.058261 1.058253 1 True 9
10 LightGBMLarge 0.828949 0.83 0.036233 0.005726 0.544085 0.036233 0.005726 0.544085 1 True 13
11 NeuralNetFastAI 0.818610 0.82 0.152624 0.013950 2.614331 0.152624 0.013950 2.614331 1 True 10
12 KNeighborsUnif 0.725970 0.73 0.027956 0.008520 0.609989 0.027956 0.008520 0.609989 1 True 1
13 KNeighborsDist 0.695158 0.65 0.025601 0.006325 0.603475 0.025601 0.006325 0.603475 1 True 2

Snapshot a Predictor with .clone()

Now that we have a working predictor artifact, we may want to alter it in a variety of ways to better suite our needs. For example, we may want to delete certain models to reduce disk usage via .delete_models(), or train additional models on top of the ones we already have via .fit_extra().

While you can do all of these operations on your predictor, you may want to be able to be able to revert to a prior state of the predictor in case something goes wrong. This is where predictor.clone() comes in.

predictor.clone() allows you to create a snapshot of the given predictor, cloning the artifacts of the predictor to a new location. You can then freely play around with the predictor and always load the earlier snapshot in case you want to undo your actions.

All you need to do to clone a predictor is specify a new directory path to clone to:

save_path_clone = save_path + '-clone'
# will return the path to the cloned predictor, identical to save_path_clone
path_clone = predictor.clone(path=save_path_clone)

Cloned TabularPredictor located in 'agModels-predictClass-deployment/' to 'agModels-predictClass-deployment-clone'.
    To load the cloned predictor: predictor_clone = TabularPredictor.load(path="agModels-predictClass-deployment-clone")

Note that this logic doubles disk usage, as it completely clones every predictor artifact on disk to make an exact replica.

Now we can load the cloned predictor:

predictor_clone = TabularPredictor.load(path=path_clone)
# You can alternatively load the cloned TabularPredictor at the time of cloning:
# predictor_clone = predictor.clone(path=save_path_clone, return_clone=True)

We can see that the cloned predictor has the same leaderboard and functionality as the original:

y_pred_clone = predictor.predict(test_data)
y_pred_clone

0        <=50K
1        <=50K
2        <=50K
3        <=50K
4        <=50K
         ...
9764     <=50K
9765     <=50K
9766     <=50K
9767     <=50K
9768     <=50K
Name: class, Length: 9769, dtype: object

y_pred.equals(y_pred_clone)

True

predictor_clone.leaderboard(test_data, silent=True)
model score_test score_val pred_time_test pred_time_val fit_time pred_time_test_marginal pred_time_val_marginal fit_time_marginal stack_level can_infer fit_order
0 RandomForestGini 0.842870 0.84 0.135722 0.055894 1.080855 0.135722 0.055894 1.080855 1 True 5
1 CatBoost 0.842461 0.85 0.011522 0.005573 1.403570 0.011522 0.005573 1.403570 1 True 7
2 RandomForestEntr 0.841130 0.83 0.134074 0.060857 1.060027 0.134074 0.060857 1.060027 1 True 6
3 LightGBM 0.839799 0.85 0.015382 0.008039 0.824368 0.015382 0.008039 0.824368 1 True 4
4 XGBoost 0.837445 0.87 0.046215 0.007187 0.261149 0.046215 0.007187 0.261149 1 True 11
5 WeightedEnsemble_L2 0.837445 0.87 0.048518 0.007834 0.583509 0.002304 0.000648 0.322360 2 True 14
6 LightGBMXT 0.836421 0.83 0.010329 0.005912 1.248788 0.010329 0.005912 1.248788 1 True 3
7 ExtraTreesGini 0.834579 0.82 0.135083 0.060351 1.065567 0.135083 0.060351 1.065567 1 True 8
8 NeuralNetTorch 0.833555 0.83 0.053613 0.013697 1.024997 0.053613 0.013697 1.024997 1 True 12
9 ExtraTreesEntr 0.833350 0.81 0.139537 0.058261 1.058253 0.139537 0.058261 1.058253 1 True 9
10 LightGBMLarge 0.828949 0.83 0.034355 0.005726 0.544085 0.034355 0.005726 0.544085 1 True 13
11 NeuralNetFastAI 0.818610 0.82 0.143432 0.013950 2.614331 0.143432 0.013950 2.614331 1 True 10
12 KNeighborsUnif 0.725970 0.73 0.026624 0.008520 0.609989 0.026624 0.008520 0.609989 1 True 1
13 KNeighborsDist 0.695158 0.65 0.025924 0.006325 0.603475 0.025924 0.006325 0.603475 1 True 2

Now let’s do some extra logic with the clone, such as calling refit_full:

predictor_clone.refit_full()

predictor_clone.leaderboard(test_data, silent=True)

Fitting 1 L1 models ...
Fitting model: KNeighborsUnif_FULL ...
    0.01s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: KNeighborsDist_FULL ...
    0.01s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: LightGBMXT_FULL ...
    0.14s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: LightGBM_FULL ...
    0.16s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: RandomForestGini_FULL ...
    0.48s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: RandomForestEntr_FULL ...
    0.47s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: CatBoost_FULL ...
    0.03s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: ExtraTreesGini_FULL ...
    0.47s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: ExtraTreesEntr_FULL ...
    0.47s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: NeuralNetFastAI_FULL ...
No improvement since epoch 0: early stopping
    0.39s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: XGBoost_FULL ...
    0.07s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: NeuralNetTorch_FULL ...
    0.56s    = Training   runtime
Fitting 1 L1 models ...
Fitting model: LightGBMLarge_FULL ...
    0.22s    = Training   runtime
Fitting model: WeightedEnsemble_L2_FULL | Skipping fit via cloning parent ...
    0.32s    = Training   runtime
Updated best model to "WeightedEnsemble_L2_FULL" (Previously "WeightedEnsemble_L2"). AutoGluon will default to using "WeightedEnsemble_L2_FULL" for predict() and predict_proba().
model score_test score_val pred_time_test pred_time_val fit_time pred_time_test_marginal pred_time_val_marginal fit_time_marginal stack_level can_infer fit_order
0 CatBoost_FULL 0.842870 NaN 0.011228 NaN 0.026799 0.011228 NaN 0.026799 1 True 21
1 RandomForestGini 0.842870 0.84 0.138639 0.055894 1.080855 0.138639 0.055894 1.080855 1 True 5
2 CatBoost 0.842461 0.85 0.012043 0.005573 1.403570 0.012043 0.005573 1.403570 1 True 7
3 RandomForestEntr 0.841130 0.83 0.138774 0.060857 1.060027 0.138774 0.060857 1.060027 1 True 6
4 LightGBM_FULL 0.840823 NaN 0.017195 NaN 0.163478 0.017195 NaN 0.163478 1 True 18
5 LightGBM 0.839799 0.85 0.015824 0.008039 0.824368 0.015824 0.008039 0.824368 1 True 4
6 RandomForestGini_FULL 0.839595 NaN 0.140190 NaN 0.478390 0.140190 NaN 0.478390 1 True 19
7 RandomForestEntr_FULL 0.839185 NaN 0.138538 NaN 0.474687 0.138538 NaN 0.474687 1 True 20
8 LightGBMXT_FULL 0.837957 NaN 0.011016 NaN 0.137915 0.011016 NaN 0.137915 1 True 17
9 XGBoost 0.837445 0.87 0.048745 0.007187 0.261149 0.048745 0.007187 0.261149 1 True 11
10 WeightedEnsemble_L2 0.837445 0.87 0.051331 0.007834 0.583509 0.002586 0.000648 0.322360 2 True 14
11 LightGBMXT 0.836421 0.83 0.010284 0.005912 1.248788 0.010284 0.005912 1.248788 1 True 3
12 ExtraTreesEntr_FULL 0.835910 NaN 0.143991 NaN 0.473450 0.143991 NaN 0.473450 1 True 23
13 NeuralNetTorch_FULL 0.835091 NaN 0.058724 NaN 0.559810 0.058724 NaN 0.559810 1 True 26
14 ExtraTreesGini 0.834579 0.82 0.142700 0.060351 1.065567 0.142700 0.060351 1.065567 1 True 8
15 ExtraTreesGini_FULL 0.833862 NaN 0.141119 NaN 0.472204 0.141119 NaN 0.472204 1 True 22
16 NeuralNetTorch 0.833555 0.83 0.057129 0.013697 1.024997 0.057129 0.013697 1.024997 1 True 12
17 ExtraTreesEntr 0.833350 0.81 0.140146 0.058261 1.058253 0.140146 0.058261 1.058253 1 True 9
18 XGBoost_FULL 0.831610 NaN 0.044393 NaN 0.069248 0.044393 NaN 0.069248 1 True 25
19 WeightedEnsemble_L2_FULL 0.831610 NaN 0.047146 NaN 0.391608 0.002753 NaN 0.322360 2 True 28
20 LightGBMLarge 0.828949 0.83 0.038662 0.005726 0.544085 0.038662 0.005726 0.544085 1 True 13
21 LightGBMLarge_FULL 0.820964 NaN 0.041921 NaN 0.220074 0.041921 NaN 0.220074 1 True 27
22 NeuralNetFastAI 0.818610 0.82 0.155864 0.013950 2.614331 0.155864 0.013950 2.614331 1 True 10
23 NeuralNetFastAI_FULL 0.769270 NaN 0.151720 NaN 0.386512 0.151720 NaN 0.386512 1 True 24
24 KNeighborsUnif 0.725970 0.73 0.025264 0.008520 0.609989 0.025264 0.008520 0.609989 1 True 1
25 KNeighborsUnif_FULL 0.725151 NaN 0.023710 NaN 0.005904 0.023710 NaN 0.005904 1 True 15
26 KNeighborsDist 0.695158 0.65 0.027080 0.006325 0.603475 0.027080 0.006325 0.603475 1 True 2
27 KNeighborsDist_FULL 0.685434 NaN 0.025221 NaN 0.005437 0.025221 NaN 0.005437 1 True 16

We can see that we were able to fit additional models, but for whatever reason we may want to undo this operation.

Luckily, our original predictor is untouched!

predictor.leaderboard(test_data, silent=True)
model score_test score_val pred_time_test pred_time_val fit_time pred_time_test_marginal pred_time_val_marginal fit_time_marginal stack_level can_infer fit_order
0 RandomForestGini 0.842870 0.84 0.140122 0.055894 1.080855 0.140122 0.055894 1.080855 1 True 5
1 CatBoost 0.842461 0.85 0.011801 0.005573 1.403570 0.011801 0.005573 1.403570 1 True 7
2 RandomForestEntr 0.841130 0.83 0.139719 0.060857 1.060027 0.139719 0.060857 1.060027 1 True 6
3 LightGBM 0.839799 0.85 0.016043 0.008039 0.824368 0.016043 0.008039 0.824368 1 True 4
4 XGBoost 0.837445 0.87 0.049586 0.007187 0.261149 0.049586 0.007187 0.261149 1 True 11
5 WeightedEnsemble_L2 0.837445 0.87 0.052166 0.007834 0.583509 0.002579 0.000648 0.322360 2 True 14
6 LightGBMXT 0.836421 0.83 0.010703 0.005912 1.248788 0.010703 0.005912 1.248788 1 True 3
7 ExtraTreesGini 0.834579 0.82 0.140917 0.060351 1.065567 0.140917 0.060351 1.065567 1 True 8
8 NeuralNetTorch 0.833555 0.83 0.060173 0.013697 1.024997 0.060173 0.013697 1.024997 1 True 12
9 ExtraTreesEntr 0.833350 0.81 0.139170 0.058261 1.058253 0.139170 0.058261 1.058253 1 True 9
10 LightGBMLarge 0.828949 0.83 0.034843 0.005726 0.544085 0.034843 0.005726 0.544085 1 True 13
11 NeuralNetFastAI 0.818610 0.82 0.160457 0.013950 2.614331 0.160457 0.013950 2.614331 1 True 10
12 KNeighborsUnif 0.725970 0.73 0.015760 0.008520 0.609989 0.015760 0.008520 0.609989 1 True 1
13 KNeighborsDist 0.695158 0.65 0.024812 0.006325 0.603475 0.024812 0.006325 0.603475 1 True 2

We can simply clone a new predictor from our original, and we will no longer be impacted by the call to refit_full on the prior clone.

Snapshot a deployment optimized Predictor via .clone_for_deployment()

Instead of cloning an exact copy, we can instead clone a copy which has the minimal set of artifacts needed to do prediction.

Note that this optimized clone will have very limited functionality outside of calling predict and predict_proba. For example, it will be unable to train more models.

save_path_clone_opt = save_path + '-clone-opt'
# will return the path to the cloned predictor, identical to save_path_clone_opt
path_clone_opt = predictor.clone_for_deployment(path=save_path_clone_opt)

Cloned TabularPredictor located in 'agModels-predictClass-deployment/' to 'agModels-predictClass-deployment-clone-opt'.
    To load the cloned predictor: predictor_clone = TabularPredictor.load(path="agModels-predictClass-deployment-clone-opt")
Clone: Keeping minimum set of models required to predict with best model 'WeightedEnsemble_L2'...
Deleting model KNeighborsUnif. All files under agModels-predictClass-deployment-clone-opt/models/KNeighborsUnif/ will be removed.
Deleting model KNeighborsDist. All files under agModels-predictClass-deployment-clone-opt/models/KNeighborsDist/ will be removed.
Deleting model LightGBMXT. All files under agModels-predictClass-deployment-clone-opt/models/LightGBMXT/ will be removed.
Deleting model LightGBM. All files under agModels-predictClass-deployment-clone-opt/models/LightGBM/ will be removed.
Deleting model RandomForestGini. All files under agModels-predictClass-deployment-clone-opt/models/RandomForestGini/ will be removed.
Deleting model RandomForestEntr. All files under agModels-predictClass-deployment-clone-opt/models/RandomForestEntr/ will be removed.
Deleting model CatBoost. All files under agModels-predictClass-deployment-clone-opt/models/CatBoost/ will be removed.
Deleting model ExtraTreesGini. All files under agModels-predictClass-deployment-clone-opt/models/ExtraTreesGini/ will be removed.
Deleting model ExtraTreesEntr. All files under agModels-predictClass-deployment-clone-opt/models/ExtraTreesEntr/ will be removed.
Deleting model NeuralNetFastAI. All files under agModels-predictClass-deployment-clone-opt/models/NeuralNetFastAI/ will be removed.
Deleting model NeuralNetTorch. All files under agModels-predictClass-deployment-clone-opt/models/NeuralNetTorch/ will be removed.
Deleting model LightGBMLarge. All files under agModels-predictClass-deployment-clone-opt/models/LightGBMLarge/ will be removed.
Clone: Removing artifacts unnecessary for prediction. NOTE: Clone can no longer fit new models, and most functionality except for predict and predict_proba will no longer work

predictor_clone_opt = TabularPredictor.load(path=path_clone_opt)

We can see that the optimized clone still makes the same predictions:

y_pred_clone_opt = predictor_clone_opt.predict(test_data)
y_pred_clone_opt

0        <=50K
1        <=50K
2        <=50K
3        <=50K
4        <=50K
         ...
9764     <=50K
9765     <=50K
9766     <=50K
9767     <=50K
9768     <=50K
Name: class, Length: 9769, dtype: object

y_pred.equals(y_pred_clone_opt)

True

predictor_clone_opt.leaderboard(test_data, silent=True)
model score_test score_val pred_time_test pred_time_val fit_time pred_time_test_marginal pred_time_val_marginal fit_time_marginal stack_level can_infer fit_order
0 XGBoost 0.837445 0.87 0.038097 0.007187 0.261149 0.038097 0.007187 0.261149 1 True 1
1 WeightedEnsemble_L2 0.837445 0.87 0.040542 0.007834 0.583509 0.002445 0.000648 0.322360 2 True 2

We can check the disk usage of the optimized clone compared to the original:

size_original = predictor.get_size_disk()
size_opt = predictor_clone_opt.get_size_disk()
print(f'Size Original:  {size_original} bytes')
print(f'Size Optimized: {size_opt} bytes')
print(f'Optimized predictor achieved a {round((1 - (size_opt/size_original)) * 100, 1)}% reduction in disk usage.')

Size Original:  16966478 bytes
Size Optimized: 601220 bytes
Optimized predictor achieved a 96.5% reduction in disk usage.

We can also investigate the difference in the files that exist in the original and optimized predictor.

Original:

predictor.get_size_disk_per_file()

models/ExtraTreesGini/model.pkl                        4567890
models/ExtraTreesEntr/model.pkl                        4530305
models/RandomForestGini/model.pkl                      3076492
models/RandomForestEntr/model.pkl                      2949158
models/XGBoost/xgb.ubj                                  564906
models/LightGBMLarge/model.pkl                          470889
models/NeuralNetTorch/net.params                        234610
models/NeuralNetFastAI/model-internals.pkl              167374
models/LightGBM/model.pkl                               146038
models/LightGBMXT/model.pkl                              42071
models/KNeighborsDist/model.pkl                          39986
models/KNeighborsUnif/model.pkl                          39985
utils/data/X.pkl                                         27655
models/CatBoost/model.pkl                                21562
models/NeuralNetTorch/model.pkl                          18149
learner.pkl                                              10719
metadata.json                                             8632
utils/data/X_val.pkl                                      8421
models/WeightedEnsemble_L2/model.pkl                      8122
utils/data/y.pkl                                          7488
models/XGBoost/model.pkl                                  5475
models/trainer.pkl                                        5124
models/NeuralNetFastAI/model.pkl                          3352
utils/data/y_val.pkl                                      2381
models/WeightedEnsemble_L2/utils/model_template.pkl       1024
models/WeightedEnsemble_L2/utils/oof.pkl                   764
predictor.pkl                                              742
utils/attr/NeuralNetTorch/y_pred_proba_val.pkl             550
utils/attr/XGBoost/y_pred_proba_val.pkl                    550
utils/attr/NeuralNetFastAI/y_pred_proba_val.pkl            550
utils/attr/ExtraTreesEntr/y_pred_proba_val.pkl             550
utils/attr/ExtraTreesGini/y_pred_proba_val.pkl             550
utils/attr/CatBoost/y_pred_proba_val.pkl                   550
utils/attr/RandomForestEntr/y_pred_proba_val.pkl           550
utils/attr/RandomForestGini/y_pred_proba_val.pkl           550
utils/attr/LightGBM/y_pred_proba_val.pkl                   550
utils/attr/LightGBMXT/y_pred_proba_val.pkl                 550
utils/attr/KNeighborsDist/y_pred_proba_val.pkl             550
utils/attr/KNeighborsUnif/y_pred_proba_val.pkl             550
utils/attr/LightGBMLarge/y_pred_proba_val.pkl              550
__version__                                                 14
Name: size, dtype: int64

Optimized:

predictor_clone_opt.get_size_disk_per_file()

models/XGBoost/xgb.ubj                  564906
learner.pkl                              10719
metadata.json                             8632
models/WeightedEnsemble_L2/model.pkl      8286
models/XGBoost/model.pkl                  5495
models/trainer.pkl                        2426
predictor.pkl                              742
__version__                                 14
Name: size, dtype: int64

Now all that is left is to upload the optimized predictor to a centralized storage location such as S3. To use this predictor in a new machine / system, simply download the artifact to local disk and load the predictor. Ensure that when loading a predictor you use the same Python version and AutoGluon version used during training to avoid instability.