Mlflow with Helm and serve Train Model on kubernetes

Part 3: Expose example api on Local Machine

Part 1: Install Mlflow on Local Machine
Part 2: Train example model and keep in Mlflow Local Machine
Part 3: Expose example api on Local Machine
Part 4: API Transform for Model API
Part 5: Install Mlflow on GKE Cluster with helm
Part 6: Keep Model in Mlflow remote Cluster
Part 7: Serve Model API on Cluster

Expose API

• From mlflow tracking server

• Copy full path of model

• Run serve model

mlflow models serve -m mlflow-artifacts:/905199336240427925/4e9807502bad4255a1d2dcdb6272cd75/artifacts/model -p 1234

• Let test api with PostMan

    
http://127.0.0.1:1234/invocations

{"dataframe_split": {"data":[[

0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,
0.0,  0.0,  1.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  1.0,  0.0

]]}}

### RESULT ###
{
    "predictions": [
        "yes"
    ]
}

• Let change data

http://127.0.0.1:1234/invocations

{"dataframe_split": {"data":[[

0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,
0.0,  0.0,  1.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
0.0,  0.0,  0.0,  1.0

]]}}

{
    "predictions": [
        "no"
    ]
}

• Let test API with curl

curl -X POST -H "Content-Type:application/json"                     \
    --data "{\"dataframe_split\": {\"data\":[[                          \
        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,
        0.0,  0.0,  1.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  1.0,  0.0]]}}"                                      \
    http://127.0.0.1:1234/invocations | jq

{
  "predictions": [
    "yes"
  ]
}

• Have fun !!!

Note data

        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,
        0.0,  0.0,  1.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  0.0,  
        0.0,  0.0,  1.0,  0.0

Come from Transform and OneHotEncoder

First “0.0” 42 item from 42 mt.ads_country_dst

And

Last “0.0” 2 item from is_OfficeHour(yes/no)

— — — — — — — — — — — — — — — — — — — — — — — — — — — — —

Credit : TrueDigitalGroup

— — — — — — — — — — — — — — — — — — — — — — — — — — — — —