Deep learning project (Keras or tensorflow)
Budget: $250 – $750 USD
I am looking for an experienced freelancer in the field of deep learning that can provide a python script in jupyter notebook. The model can use tensorflow or keras.
The script must have the following features:
-k-fold cross validation on train and validation sets and a test set
-hyperparameter tuning (e.g. keras tuner)
-explainable ai with lime, shap and tf-explain (if you have better suggestions, we can do something else)
-confusion matrix
-a csv export of the model output for the test set which must contain for each of the cases the id and model output for that case (THIS IS VERY IMPORTANT, i will use this export later in a random forest script)
In the attachment you have part of the dataset (ECG signal) and transformation from signal to image using spectrogram. You can use this code to transform signal to image, no signal processing experience is required since you have the code for this part.
The script must have the following features:
-k-fold cross validation on train and validation sets and a test set
-hyperparameter tuning (e.g. keras tuner)
-explainable ai with lime, shap and tf-explain (if you have better suggestions, we can do something else)
-confusion matrix
-a csv export of the model output for the test set which must contain for each of the cases the id and model output for that case (THIS IS VERY IMPORTANT, i will use this export later in a random forest script)
In the attachment you have part of the dataset (ECG signal) and transformation from signal to image using spectrogram. You can use this code to transform signal to image, no signal processing experience is required since you have the code for this part.