Update README.md

README.md

@@ -31,7 +31,7 @@ We also include `gender_submission.csv`, a set of predictions that assume all an
 ### Problem Statement
 You can choose to do one or more of the following tasks - 
 1. **Exploratory Data Visualisation** - The notebooks already have the part included which provide analysis of main characteristics of data. You are welcome to add to this for visualising more trends.
-2. **Build a Grader** - An `output.csv` file has been included in the folder with [titanic dataset](./titanic). It has the expected predictions corresponding to [test data](./titanic/test.csv) and your task is to write a Python fucntion which takes as input a csv file, compares to the expected output and prints the percentage of accuracy achieved.
+2. **Build a Grader** - An `output.csv` file has been included in the folder with [titanic dataset](./titanic). It has the expected predictions corresponding to [test data](./titanic/test.csv) and your task is to write a Python function which takes as input a csv file, compares to the expected output and prints the percentage of accuracy achieved.
 3. **Hyperparameter tuning** - A neural network has been trained on the dataset in [this notebook](./Titanic%20neural%20network%20Tensorflow.ipynb) but it has very low accuracy. Your task is to tune the hyperparameters and improve the model. Some suggestions - modifying the optimizer, adding more hidden layers to the model or changing the dimension of layers, adding dropout, regularization etc.
 4. **Back-Propagation in Numpy** - A neural network has been coded from scratch in Numpy on the dataset in [this notebook](./Titanic%20neural%20network%20Numpy%20from%20scratch.ipynb) but it is incomplete; it misses the function for calculating gradients during back-propagation. Your task is to complete the function for calculating gradients, train the model and output the predictions on test data to a csv file.
 5. **Implementing algorithm in sklearn** - Apart from a neural network, there are many other ML algorithms that can be used to make predictions in this challenge. Notebooks for some of them have been put up but only with data reading and visualisation part completed. Your task is to write the code for implementing the specificied algortihm on the dataset using sklearn library and printing the predictions to a csv file. You can choose from any of the following or get started with one of your own!