Build an ML model for meCristian Popa
An ML model consists of a set of files that, after being trained on some examples and with the help of some code, is able to make predictions on unseen data. This has applications for a lot of different tasks:
Classification (e.g. predicting a disease based on images of patients)
Predicting numbers (e.g. the price of a house in a certain area)
Forecasting (e.g. the price of houses after some piece of news comes out)
Some ML models don't need training data and can solve a problem in a different way, but it depends on the issue.
Important note: It is difficult to correctly estimate the performance of an ML model prior to building it. For this reason, I can only offer rough estimations of expected performance citing similar work, but the model achieving that performance is not guaranteed. I will provide an analysis of all the performance metrics. The experimentation stops when you are content with the results or I run out of ideas.
What's included
ML model
This includes any files generated after training the ML model. These files are used to load the model on your own machine and use it to make predictions on the task.
Code
This includes code for all the applicable steps in the pipeline:
- Data gathering
- Data processing
- Model training (and fine-tuning, if applicable)
- Model inference (prediction) and evaluation
Data (optional)
If the data had to be gathered and I handled it, you will get all of it (in a processed form, if it is more convenient).
Performance Breakdown
The results obtained by the different models trained in an eye-candy way - I use Streamlit usually.
Documentation
A detailed README that describes everything you would need to use the other deliverables:
- Where the model and data is
- How to run the code
- Small summary of the code and its different components
Cristian's other services
Starting at$50 /hr
Tags
Docker
Python
PyTorch
Data Scientist
ML Engineer
Service provided by
Cristian Popa Bucharest, Romania
Build an ML model for meCristian Popa
Starting at$50 /hr
Tags
Docker
Python
PyTorch
Data Scientist
ML Engineer
An ML model consists of a set of files that, after being trained on some examples and with the help of some code, is able to make predictions on unseen data. This has applications for a lot of different tasks:
Classification (e.g. predicting a disease based on images of patients)
Predicting numbers (e.g. the price of a house in a certain area)
Forecasting (e.g. the price of houses after some piece of news comes out)
Some ML models don't need training data and can solve a problem in a different way, but it depends on the issue.
Important note: It is difficult to correctly estimate the performance of an ML model prior to building it. For this reason, I can only offer rough estimations of expected performance citing similar work, but the model achieving that performance is not guaranteed. I will provide an analysis of all the performance metrics. The experimentation stops when you are content with the results or I run out of ideas.
What's included
ML model
This includes any files generated after training the ML model. These files are used to load the model on your own machine and use it to make predictions on the task.
Code
This includes code for all the applicable steps in the pipeline:
- Data gathering
- Data processing
- Model training (and fine-tuning, if applicable)
- Model inference (prediction) and evaluation
Data (optional)
If the data had to be gathered and I handled it, you will get all of it (in a processed form, if it is more convenient).
Performance Breakdown
The results obtained by the different models trained in an eye-candy way - I use Streamlit usually.
Documentation
A detailed README that describes everything you would need to use the other deliverables:
- Where the model and data is
- How to run the code
- Small summary of the code and its different components
Cristian's other services
$50 /hr