School 42 Data Science Project
Code Hogwarts school's Sorting Hat
python3 describe.py dataset_path
- Display features informations
python3 histogram.py dataset_path
- Answer the following question: Which Hogwarts course has a homogeneous score distribution between the four houses ?
python3 scatter_plot.py dataset_path
- Answer the following question: Which features are similar ?
python3 pair_plot.py dataset_path
- Pair_plot of all features
logreg_train.py [-h] [-s] [-a] [-c] [-p] [-f] [-e E] [-ls] [-log] [-v] file
positional arguments: file input dataset path
optional arguments:
-h, --help show this help message and exit
-s Stochastic gradient descent
-a Stochastic gradient descent with adam optimizer
-c Check results
-p Plot cost function
-f Forward fill method for NaN values. Default: DropNaN
-e int Epochs Number. Default: 100
-ls Linear Scalling. Default : Z_score
-log Log Scalling. Default : Z_score
-v Show compare
python3 logreg_predict.py dataset weights
- Generate a file with all predictions for a given dataset
###Example :
logreg_train.py -c -e 100 file