Model Deployment

Bluemist provides integration with FastAPI to deploy the ML model as RESTful API. This can be simply achieved by calling deploy_model as described here

Warning

deploy_model should only be executed after train_test_evaluate has been completed successfully

Examples

regression_model_deployment

In [1]:

pip install bluemist

In [2]:

import pandas as pd

from bluemist.environment import initialize
from bluemist.preprocessing import preprocess_data
from bluemist.regression import train_test_evaluate, deploy_model

In [3]:

initialize()
data = pd.read_csv('https://www4.stat.ncsu.edu/~boos/var.select/diabetes.tab.txt', sep='\t')
data.head()


██████╗ ██╗     ██╗   ██╗███████╗███╗   ███╗██╗███████╗████████╗     █████╗ ██╗
██╔══██╗██║     ██║   ██║██╔════╝████╗ ████║██║██╔════╝╚══██╔══╝    ██╔══██╗██║
██████╔╝██║     ██║   ██║█████╗  ██╔████╔██║██║███████╗   ██║       ███████║██║
██╔══██╗██║     ██║   ██║██╔══╝  ██║╚██╔╝██║██║╚════██║   ██║       ██╔══██║██║
██████╔╝███████╗╚██████╔╝███████╗██║ ╚═╝ ██║██║███████║   ██║       ██║  ██║██║                                                                        
                                (version 0.1.1)
    
Bluemist path :: /home/shashank-agrawal/PycharmProjects/bluemist-ai/bluemist
System platform :: posix, Linux, 5.19.0-31-generic, linux-x86_64, ('64bit', 'ELF')

Out[3]:

	AGE	SEX	BMI	BP	S1	S2	S3	S4	S5	S6	Y
0	59	2	32.1	101.0	157	93.2	38.0	4.0	4.8598	87	151
1	48	1	21.6	87.0	183	103.2	70.0	3.0	3.8918	69	75
2	72	2	30.5	93.0	156	93.6	41.0	4.0	4.6728	85	141
3	24	1	25.3	84.0	198	131.4	40.0	5.0	4.8903	89	206
4	50	1	23.0	101.0	192	125.4	52.0	4.0	4.2905	80	135

In [4]:

# Categorical encoding using OneHotEncoder
X_train, X_test, y_train, y_test = preprocess_data(data, 
                                                   target_variable='Y', 
                                                   test_size=0.25, 
                                                   data_scaling_strategy='StandardScaler',
                                                   categorical_features=['SEX'], 
                                                   categorical_encoder='OneHotEncoder',
                                                   drop_categories_one_hot_encoder='first')
X_train.head()

Out[4]:

	AGE	BMI	BP	S1	S2	S3	S4	S5	S6	SEX_2
0	-2.256402	-0.237583	-0.834117	0.975180	1.289605	-0.302383	0.687862	0.125787	-0.668411	0.0
1	1.230453	1.618094	0.520919	-0.006013	0.320029	-1.158176	0.687862	0.605781	0.916937	1.0
2	-1.801595	-0.644927	-0.477529	-1.323615	-1.127958	0.397811	-0.818322	-2.042113	-0.756486	1.0
3	0.320838	1.527573	1.590684	-0.202252	0.090392	-0.146785	-0.065230	-0.516943	1.269236	1.0
4	0.624043	1.188120	1.590684	0.975180	-0.273199	-0.769179	1.440954	2.494051	1.885760	1.0

In [5]:

# Train and compare models
train_test_evaluate(X_train, X_test, y_train, y_test, target_scaling_strategy='MinMaxScaler')

Training TweedieRegressor: 100%|██████████| 46/46 [00:20<00:00,  2.25it/s]

	mean_absolute_error	mean_squared_error	r2_score	Exception
Estimator
ARDRegression	44.905751	3132.297360	0.454205	nan
AdaBoostRegressor	48.719321	3818.411711	0.334652	nan
BaggingRegressor	50.486486	4114.979459	0.282976	nan
BayesianRidge	45.401862	3169.959732	0.447643	nan
CCA	55.569805	4685.287437	0.183601	nan
DecisionTreeRegressor	71.225225	8269.369369	-0.440916	nan
DummyRegressor	64.621513	5793.204870	-0.009451	nan
ElasticNet	64.621513	5793.204870	-0.009451	nan
ElasticNetCV	45.415690	3186.492960	0.444762	nan
ExtraTreeRegressor	67.666667	7536.837838	-0.313274	nan
ExtraTreesRegressor	45.631892	3580.983005	0.376023	nan
GammaRegressor	nan	nan	nan	Some value(s) of y are out of the valid range of the loss 'HalfGammaLoss'.
GaussianProcessRegressor	67.216723	8630.149234	-0.503781	nan
GradientBoostingRegressor	48.374687	3934.861451	0.314361	nan
HistGradientBoostingRegressor	48.686275	4010.794382	0.301130	nan
HuberRegressor	44.761295	3126.524871	0.455211	nan
KNeighborsRegressor	52.052252	4569.992432	0.203691	nan
KernelRidge	93.436868	13826.460935	-1.409225	nan
Lars	45.033817	3118.568339	0.456598	nan
LarsCV	45.606204	3222.114649	0.438555	nan
Lasso	64.621513	5793.204870	-0.009451	nan
LassoCV	45.512171	3209.088449	0.440825	nan
LassoLars	64.621513	5793.204870	-0.009451	nan
LassoLarsCV	45.606204	3222.114649	0.438555	nan
LassoLarsIC	45.421964	3191.847968	0.443829	nan
LinearRegression	45.033817	3118.568339	0.456598	nan
LinearSVR	44.688204	3056.343818	0.467440	nan
MLPRegressor	51.895604	4568.649034	0.203925	nan
NuSVR	48.338294	4112.313615	0.283440	nan
OrthogonalMatchingPursuit	52.597461	4691.442401	0.182528	nan
OrthogonalMatchingPursuitCV	45.962764	3298.233809	0.425291	nan
PLSCanonical	90.688422	11521.103125	-1.007522	nan
PLSRegression	45.026657	3129.231385	0.454740	nan
PassiveAggressiveRegressor	62.835621	6030.485222	-0.050796	nan
PoissonRegressor	52.590634	4032.492444	0.297349	nan
QuantileRegressor	64.351351	6050.441443	-0.054274	nan
RANSACRegressor	51.590759	4285.951082	0.253184	nan
RadiusNeighborsRegressor	49.636364	4217.090909	0.324830	nan
RandomForestRegressor	47.074234	3831.500310	0.332371	nan
Ridge	45.110837	3130.333945	0.454547	nan
RidgeCV	45.336901	3161.739093	0.449075	nan
SGDRegressor	47.938802	3587.937037	0.374811	nan
SVR	49.777276	4135.553245	0.279391	nan
TheilSenRegressor	46.437096	3422.785127	0.403589	nan
TransformedTargetRegressor	45.033817	3118.568339	0.456598	nan
TweedieRegressor	48.923752	3586.863151	0.374998	nan

In [ ]:

deploy_model(estimator_name='LinearSVR', host='localhost', port=8000)

INFO:     Started server process [167513]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://localhost:8000 (Press CTRL+C to quit)

INFO:     127.0.0.1:40194 - "GET / HTTP/1.1" 404 Not Found
INFO:     127.0.0.1:40194 - "GET /docs HTTP/1.1" 200 OK
INFO:     127.0.0.1:40194 - "GET /openapi.json HTTP/1.1" 200 OK
INFO:     127.0.0.1:40210 - "POST /predict HTTP/1.1" 200 OK

To test the API, open the browser and navigate to http://localhost:8000/docs