Flask API for model serving in ML Python - Model Pipeline Trace

Model Pipeline - Flask API for model serving

This pipeline shows how a trained machine learning model is served using a Flask API. The API receives input data, processes it, uses the model to predict, and returns the prediction as a response.

Data Flow - 4 Stages

1Input Data Received

1 row x 4 features→Receive JSON data from API request→1 row x 4 features

{"sepal_length": 5.1, "sepal_width": 3.5, "petal_length": 1.4, "petal_width": 0.2}

↓

2Data Preprocessing

1 row x 4 features→Convert JSON to numpy array and scale features→1 row x 4 features

[[0.22, -0.42, 0.11, -0.56]]

↓

3Model Prediction

1 row x 4 features→Feed preprocessed data into trained model→1 row x 3 classes

[[0.95, 0.03, 0.02]]

↓

4Output Response

1 row x 3 classes→Convert prediction probabilities to class label and send JSON response→JSON response

{"predicted_class": "setosa"}

Training Trace - Epoch by Epoch

Loss
0.7 |****
0.6 |*** 
0.5 |**  
0.4 |**  
0.3 |*   
0.2 |*   
0.1 |    
    +------------
     1 2 3 4 5 Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	0.65	0.70	Model starts learning with moderate accuracy.
2	0.45	0.85	Loss decreases and accuracy improves significantly.
3	0.30	0.92	Model converges with high accuracy.
4	0.25	0.95	Further fine-tuning improves performance.
5	0.20	0.97	Training stabilizes with low loss and high accuracy.

Prediction Trace - 4 Layers

Layer 1: API Input Parsing

Layer 2: Feature Scaling

Layer 3: Model Prediction

Layer 4: Response Formatting

Model Quiz - 3 Questions

What shape does the input data have when the API first receives it?

A 1 row x 4 features

B 100 rows x 4 features

C 1 row x 3 classes

D 4 rows x 1 feature

Key Insight

Serving a trained model with Flask involves receiving input data, preprocessing it exactly as during training, predicting with the model, and returning clear results. This pipeline ensures consistent and reliable predictions in real-time applications.