diff --git a/README.md b/README.md
index b5070be..7517aea 100644
--- a/README.md
+++ b/README.md
@@ -13,16 +13,16 @@ We are presenting a CNN approach using two convolutional networks to classify hi
 
 ## Getting Started
 ### Installation
-- Install PyTorch and dependencies from http://pytorch.org
-- Install python scikit-learn libraries [scikit-learn](https://github.com/scikit-learn/scikit-learn).
-```bash
-pip install -U scikit-learn
-```
 - Clone this repo:
 ```bash
 git clone https://github.com/ImagingLab/ICIAR2018
 cd ICIAR2018
 ```
+- Install PyTorch and dependencies from http://pytorch.org
+- Install python requirements:
+```bash
+pip install -r requirements.txt
+```
 
 ### Dataset
 - We use the ICIAR2018 dataset. To train a model on the full dataset, please download it from the [official website](https://iciar2018-challenge.grand-challenge.org/dataset/) (registration required). The dataset is composed of 400 high resolution Hematoxylin and Eosin (H&E) stained breast histology microscopy images labelled as normal, benign, in situ carcinoma, and invasive carcinoma (100 images for each category):
diff --git a/requirements.txt b/requirements.txt
index bfdddea..bafdf95 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,6 +1,6 @@
-numpy==1.14.3
-scipy==1.0.1
-future==0.16.0
-matplotlib=2.2.2
-pillow==5.0.0
-scikit-learn==0.19.1
\ No newline at end of file
+numpy ~=1.14.3
+scipy ~= 1.0.1
+future ~= 0.16.0
+matplotlib ~= 2.2.2
+pillow != 5.0.0
+scikit-learn ~= 0.19.1