python - Fever forecasting with multiple patients for each time step with LSTM

Question

I would like to predict fever temperature for the next hours of the week, for each of the patient in my dataset. I was thinking of using LSTMs for achieving this.

The dataset has multiple data inputs for each time splot

Each patient has multiple features for each day and gives an output in the range of [30, 43]. Not all users have data for each day.

-----------------------------------------------------------------
|     day    |   patient_id   | feature_1 |   feature_n  |  output |
-----------------------------------------------------------------
| 2020-02-11 |      1      |     20    |       9      |    38    |
| 2020-02-11 |      2      |     15    |       8      |    34    |
| 2020-02-12 |      1      |     10    |       2      |    36    |
| 2020-02-13 |      2      |     16    |       9      |    35    |
| 2020-02-13 |      3      |     19    |       1      |    37    |
| 2020-02-13 |      1      |     14    |       4      |    39    |

I want to forecast the output for each of the patient for the next hours of the week.

I was thinking to split the dataset in multiple datasets for each of the patients and use only the data for the patient to predict the output. What do you think about this idea? is there any other better idea (cause i have thousands of patient, so i will have at final thousands csv file)

Is there any way to use the whole dataset and after training the model to get as output the output value for each of the patients for the next hours of the week?

question from:https://stackoverflow.com/questions/66050549/fever-forecasting-with-multiple-patients-for-each-time-step-with-lstm

Answer 1

Waitting for answers