diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..6794eec
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,31 @@
+FROM python:3.10-slim
+
+LABEL author="Terézia Slanináková"
+LABEL email="slaninakova@mail.muni.cz"
+LABEL website="https://disa.fi.muni.cz/complex-data-analysis/"
+
+# Set the version of the image to use, default: cpu
+ARG version=cpu
+
+# Install linux packages
+RUN apt-get update && apt-get install vim -y
+
+# Install required python packages
+COPY requirements-${version}.txt /tmp/
+COPY requirements.txt /tmp/
+RUN pip install --upgrade pip
+RUN pip install --no-cache-dir -r /tmp/requirements-${version}.txt
+
+# Create user, make it the owner of the home directory
+RUN addgroup --gid 1000 user && adduser --gid 1000 --uid 1000 --disabled-password --gecos user user
+USER root
+RUN chown -R user:user /home/user && chmod -R 755 /home/user
+
+# Copy the files from the host to the container and install the local package
+COPY . /home/user
+RUN pip install -e /home/user
+
+USER user
+WORKDIR /home/user
+
+CMD ['/bin/sh', '-c', 'bash']
\ No newline at end of file
diff --git a/README.md b/README.md
index 22eb69f..8202bff 100644
--- a/README.md
+++ b/README.md
@@ -2,40 +2,49 @@
 
 Learned Metric Index (LMI) is an index for approximate nearest neighbor search on complex data using machine learning and probability-based navigation. 
 
-
 # Getting started
 
 See examples of how to index and search in a dataset in: [01_Introduction.ipynb](01_Introduction.ipynb) notebook.
 
 ## Installation
 
-See also ``.github/workflows/ci.yml``
-
-### Using conda
+### Using virtualenv
 ```bash
-conda create -n env python=3.8
-conda activate env
-conda install matplotlib pandas scikit-learn jupyterlab
-pip install h5py flake8 setuptools tqdm faiss-cpu
-pip install torch --index-url https://download.pytorch.org/whl/cpu
+# 1) Clone the repo with submodules 
+git clone --recursive git@github.com:LearnedMetricIndex/LearnedMetricIndex.git
+# 2) Create and activate a new virtual environment
+python -m venv lmi-env
+source lmi-env/bin/activate
+# 3) Install the dependencies
+pip install -r requirements-cpu.txt # alternatively requirements-gpu.txt
 pip install --editable .
 ```
 
-## Running
+### Using docker
 
-```bash
-jupyter-lab
-# and open 01_Introduction.ipynb
+Requirements:
+- [Docker](https://docs.docker.com/get-docker/)
+- At least 1.5 gb disk space for the CPU and up to 5.5 gb for the GPU version
 
-# or
-python3 search/search.py
+```bash
+# 1) Clone the repo with submodules 
+git clone --recursive git@github.com:LearnedMetricIndex/LearnedMetricIndex.git
+# 2) Build the docker image (CPU version)
+docker build -t lmi -f Dockerfile --build-arg version=cpu .
+# alternatively: docker build -t lmi -f Dockerfile --build-arg version=gpu .
+# 3) Run the docker image
+docker run -p 8888:8888 -it lmi bash
 ```
 
-## Evaluation
+## Running
 
 ```bash
-python3 eval/eval.py
-python3 eval/plot.py res.csv
+# Run jupyterlab, copy the outputted url into the browser and open 01_Introduction.ipynb
+jupyter-lab --ip 0.0.0.0 --no-browser
+
+# Run the search on 100k data subset, evaluate the results and plot them.
+# Expected time to run = ~5-10 mins
+python3 search/search.py && python eval/eval.py && python eval/plot.py res.csv
 ```
 
 ## Performance
@@ -64,7 +73,8 @@ python3 eval/plot.py res.csv
 - ~6h of runtime (waries depending on the hardware)
 
 # LMI in action
-🌐 [**Similarity search in 214M protein structures (AlphaFold DB)**](https://alphafind.fi.muni.cz/search)
+
+- 🌐 [**Similarity search in 214M protein structures (AlphaFold DB)**](https://alphafind.fi.muni.cz/)
 
 # Publications
 
@@ -86,7 +96,7 @@ python3 eval/plot.py res.csv
 - [**Web**](https://alphafind.fi.muni.cz/search)
 - [**Repository**](https://github.com/Coda-Research-Group/AlphaFind)
 - [**Data**](https://data.narodni-repozitar.cz/general/datasets/d35zf-1ja47)
-> PROCHÁZKA, David, Terézia SLANINÁKOVÁ, Jaroslav OĽHA, Adrián ROŠINEC, Katarína GREŠOVÁ, Miriama JÁNOŠOVÁ, Jakub ČILLÍK, Jana PORUBSKÁ, Radka SVOBODOVÁ, Vlastislav DOHNAL a Matej ANTOL.: [AlphaFind: Discover structure similarity across the entire known proteome](https://www.biorxiv.org/content/10.1101/2024.02.15.580465v1). BioRxiv (pre-print version)
+> Procházka, D., Slanináková, T., Oľha, J., Rošinec, A., Grešová, K., Jánošová, M., Čillík, J., Porubská, J., Svobodová, R., Dohnal, V., & Antol, M. (2024). [AlphaFind: discover structure similarity across the proteome in AlphaFold DB](https://academic.oup.com/nar/article/52/W1/W182/7673488). Nucleic Acids Research.
 
 
 ## Team
diff --git a/requirements-cpu.txt b/requirements-cpu.txt
new file mode 100644
index 0000000..bdb681f
--- /dev/null
+++ b/requirements-cpu.txt
@@ -0,0 +1,3 @@
+-r requirements.txt
+--extra-index-url=https://download.pytorch.org/whl/cpu
+torch==2.1.1
\ No newline at end of file
diff --git a/requirements-gpu.txt b/requirements-gpu.txt
new file mode 100644
index 0000000..6d914b2
--- /dev/null
+++ b/requirements-gpu.txt
@@ -0,0 +1,2 @@
+-r requirements.txt
+torch==2.1.1
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..da0b0c3
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,10 @@
+numpy==1.26.2
+matplotlib==3.8.1
+pandas==2.1.3
+scikit-learn==1.3.2
+h5py==3.10.0
+flake8==6.1.0
+tqdm==4.66.1
+faiss-cpu==1.7.4
+setuptools==60.7.0
+jupyterlab==4.0.8
\ No newline at end of file
diff --git a/search/li/clustering/faiss_kmeans.py b/search/li/clustering/faiss_kmeans.py
index 73c3c70..6026b76 100644
--- a/search/li/clustering/faiss_kmeans.py
+++ b/search/li/clustering/faiss_kmeans.py
@@ -16,6 +16,7 @@ def cluster(
     _, d = data.shape
 
     kmeans = Kmeans(d=d, k=n_clusters, **parameters)
+    data = np.ascontiguousarray(data.astype(np.float32))
     kmeans.train(data)
 
     labels = kmeans.index.search(data, 1)[1].T[0]  # type: ignore