Train with python. Predict with C++
Machine Learning everywhere!
Dream team
Dream team – synergy way
Dream team – process way
Machine learning sample cases
Buildings Energy Efficiency
Regression problem
Regression problem
Regression problem
Quality metric
Baseline model
Linear regression
Polynomial regression
Polynomial regression
Integration testing
Intrusion detection system
Classification problem
Quality metrics
Baseline model
Logistic regression
Logistic regression
Gradient boosting
CatBoost
CatBoost
Image classification
Multilayer perceptron
Quality metrics
Multilayer perceptron
Convolutional networks
Tensorflow
Conclusion
References
Let’s Talk?

Train with python. Predict with C++

1. Train with python. Predict with C++

Pavel Filonov, C++ Siberia 2019

2. Machine Learning everywhere!

• Mobile
• Embedded
• Automotive
• Desktops
• Games
• Finance
• Etc.
22
Image from [1]

3. Dream team

Developer
3
Data Scientist

4. Dream team – synergy way

Developer
4
Data Scientist
Research Developer

5. Dream team – process way

Communications
Developer
5
Data Scientist

6. Machine learning sample cases

1. Energy efficiency prediction
2. Intrusion detection system
3. Image classification
6

7. Buildings Energy Efficiency

• Input attributes
• Relative Compactness
• Surface Area
• Wall Area
• etc.
• Outcomes
• Heating Load
7
ref: [2]

8. Regression problem

8

9. Regression problem

9

10. Regression problem

10

11. Quality metric

• Determination coefficient
English     Русский Правила