beginner-level Machine Learning project
Budget: $30 – $250 USD
I have a request for a simple implementation of a Neural Network targeting two fields: Latitude, Longitude. I have attached the data.json file for you to use. The goal of this implementation is to predict the Latitude and Longitude based on the data provided in the attached file.
Here are the details for the implementation:
The Neural Network should be implemented using Python and its popular deep learning libraries such as Keras, TensorFlow or PyTorch.
The data provided in the attached data.json file should be used to train the Neural Network. The data should be split into training and test sets to evaluate the performance of the model.
The Neural Network should be designed to take a set of inputs (same as data.json) and make predictions for Latitude and Longitude. The inputs and outputs should be normalized to ensure that the model can make accurate predictions.
The Neural Network should be optimized using an appropriate optimization algorithm such as Stochastic Gradient Descent (SGD) or Adaptive Moment Estimation (Adam).
The performance of the Neural Network should be evaluated using appropriate evaluation metrics such as Mean Squared Error (MSE) or Root Mean Squared Error (RMSE).
Once the Neural Network has been trained and its performance has been evaluated, it should be tested on an out-of-sample final set to ensure that it generalizes well to unseen data.
ADDITIONAL DETAILS
First, let me clarify the purpose of the Neural Network. We are trying to predict two fields: Latitude and Longitude based on the data that we have in the data.json file. So the goal is to use the data in this file to train the Neural Network so that it can make accurate predictions for these two fields.
Second, regarding the training set, test set, and out-of-sample final set - these are just different subsets of the data that we use for different purposes. The training set is used to actually train the Neural Network and make it better at making predictions. The test set is used to see how well the Neural Network is performing on data it hasn't seen before. The out-of-sample final set is used to get an overall idea of how well the Neural Network is performing on new, unseen data (use the entries with missing values of latitude/longitude for this).
So, to summarize:
Use the data in data.json to train the Neural Network to make accurate predictions for Latitude and Longitude
Divide the data into a training set, test set, and out-of-sample final set for the purposes described above.
I have attached some PDF files with additional details, and also the field definitions (schema.py) and some pseudo-code of the solution I want.
Here are the details for the implementation:
The Neural Network should be implemented using Python and its popular deep learning libraries such as Keras, TensorFlow or PyTorch.
The data provided in the attached data.json file should be used to train the Neural Network. The data should be split into training and test sets to evaluate the performance of the model.
The Neural Network should be designed to take a set of inputs (same as data.json) and make predictions for Latitude and Longitude. The inputs and outputs should be normalized to ensure that the model can make accurate predictions.
The Neural Network should be optimized using an appropriate optimization algorithm such as Stochastic Gradient Descent (SGD) or Adaptive Moment Estimation (Adam).
The performance of the Neural Network should be evaluated using appropriate evaluation metrics such as Mean Squared Error (MSE) or Root Mean Squared Error (RMSE).
Once the Neural Network has been trained and its performance has been evaluated, it should be tested on an out-of-sample final set to ensure that it generalizes well to unseen data.
ADDITIONAL DETAILS
First, let me clarify the purpose of the Neural Network. We are trying to predict two fields: Latitude and Longitude based on the data that we have in the data.json file. So the goal is to use the data in this file to train the Neural Network so that it can make accurate predictions for these two fields.
Second, regarding the training set, test set, and out-of-sample final set - these are just different subsets of the data that we use for different purposes. The training set is used to actually train the Neural Network and make it better at making predictions. The test set is used to see how well the Neural Network is performing on data it hasn't seen before. The out-of-sample final set is used to get an overall idea of how well the Neural Network is performing on new, unseen data (use the entries with missing values of latitude/longitude for this).
So, to summarize:
Use the data in data.json to train the Neural Network to make accurate predictions for Latitude and Longitude
Divide the data into a training set, test set, and out-of-sample final set for the purposes described above.
I have attached some PDF files with additional details, and also the field definitions (schema.py) and some pseudo-code of the solution I want.