comments, readme(to be rechek)

.gitignore

@@ -7,4 +7,13 @@ dataset/train.csv
 dataset/*
 src/old_models/
 config/
-.idea/
+.idea/
+src/albi_test.py
+src/helper.py
+src/__pycache__/ensemble_log_regression.cpython-36.pyc
+src/__pycache__/helper.cpython-36.pyc
+src/dataset/
+.idea/workspace.xml
+src/__pycache__/utils.cpython-36.pyc
+
+

README.md

@@ -4,12 +4,13 @@ The model developed is able to classify the events (signal or noise)
 relevant for the detection of the Higgs Boson. More info in https://www.kaggle.com/c/higgs-boson/leaderboard
 ## Getting Started
 Git clone the repository.
+https://github.com/Idate96/cs_433_ML_project_1.git
 To run the model:
 ```
 python run.py
 ```
 It will reproduce the best results obtained with it.
-They are saved in the folder results.
+They are saved in the folder dataset under the name submission_0x.csv.
 
 ### Prerequisites
 ```
@@ -19,7 +20,23 @@ numpy  1.13.x
 
 ## Running the tests
 ```
-nosetests --verbose
+nosetests --verbose Do we need this?
+```
+
+## Architecture of the code
+
+```
+
+-src
+--run.py # main
+--ensemble_log_regression.py # contains the regression model
+--utils.py # helper function to process and load data
+--implementations.py # contains the implementation of ML_methods
+--test_ML_methods # contains the dataset used during the lab sessions
+--old_models # old_models considered for the assignment
+-dataset # contains the dataset for the competition and the submission for kaggle
+-config #where the weight of the model are saved
+
 ```
 
 ## Authors

src/ensemble_log_regression.py

@@ -1,3 +1,6 @@
+"""
+Contains the Config object, LogisticClassifier object and EnsembleClassifiers object
+"""
 import sys
 import os
 import matplotlib.pyplot as plt
@@ -12,7 +15,7 @@
 
 class Config(object):
     """Configuration object for the classifiers
-    batch sixe, number of epochs (the amount of time the program goes through de dataset, learning rate is the step
+    batch size, number of epochs (the amount of time the program goes through de dataset, learning rate is the step
     of the gradient, the lambda and mode can be either cross-validation or test"""
 
     def __init__(self, batch_size, num_epochs, learning_rate, lambda_, mode='cv'):