Technology
Artificial Intelligence
Data Analysis
Machine Learning
Data Cleaning
Model Training
Model Evaluation
Model Deployment
A Typical Machine Learning Workflow
1. Problem formulation
2. Data collection
3. Data analysis
4. Data cleaning
5. Selecting and training a model
6. Evaluating a model
7. Improving a model/data
8. Evaluating the model (on test)
9. Deploying the model
1. Problem formulation
2. Data collection
3. Data analysis
4. Data cleaning
5. Selecting and training a model
6. Evaluating a model
7. Improving a model/data
8. Evaluating the model (on test)
9. Deploying the model
As much as I can, I try to structure all my ML works based off the above workflow and it helps a lot.
If I understand the problem well, I know what to look in the data. If I analyze the data well, I know the proper data cleaning techniques.....
If I understand the problem well, I know what to look in the data. If I analyze the data well, I know the proper data cleaning techniques.....
If I cleaned the data well, I won't have issues with models input errors or so. If I evaluated the model well, I know if I have to improve the model or the data. And I know I should never provide the test data to the model until I improve it.
Loading suggestions...