Add Docker build

.dockerignore

@@ -0,0 +1,7 @@
+.git
+.tox
+.vscode
+build
+*cache*
+*/data
+*env*

Dockerfile

@@ -0,0 +1,22 @@
+FROM python:3.6-slim-stretch
+RUN apt-get update \
+    && apt-get install -y \
+    apt-utils \
+    build-essential \
+    git \
+    wget \
+    curl \
+    unzip
+
+WORKDIR /app
+COPY . /app
+ARG TUTORIAL
+RUN wget -O snorkel-requirements.txt \
+    https://raw.githubusercontent.com/HazyResearch/snorkel/redux/requirements.txt \
+    && pip3 install -r $TUTORIAL/requirements.txt \
+    && pip3 install -r requirements.txt \
+    && pip3 install -r snorkel-requirements.txt \
+    && python3 -m spacy download en_core_web_sm
+
+WORKDIR $TUTORIAL
+ENTRYPOINT ["jupyter", "notebook", "--ip=0.0.0.0", "--no-browser", "--allow-root"]

README.md

@@ -15,25 +15,55 @@ We recommend that all users **start with the `spam` tutorial** for a gentle intr
 All other tutorials assume that you have already completed that tutorial and are familiar with its concepts.
 
 ## Getting Started
-Running a tutorial has four required steps:
 
-1. Cloning this repo
+Step one is cloning this repo.
+If you're looking to quickly get started with a tutorial, we recommend using our [Docker setup](#docker).
+If you want to install things yourself, you can follow our [installation steps](#install) below instead.
+
+### <a name="docker"> Running with Docker </a>
+
+We've included a Docker setup for our tutorials to make setup easy.
+First, make sure you have [Docker installed](https://docs.docker.com/install/) on your machine.
+To build and run a Docker image for a tutorial, use `scripts/docker_launch.py` with the `--build` flag.
+For example, run the following for the `spam` tutorial:
+
+```bash
+python3 scripts/docker_launch.py spam --build
+```
+
+This will make a Jupyter notebook server available on port 8888
+(you can change the port with the `--port` command line option)
+and print out a link you can follow to access the browser interface.
+In your browser, open a `.ipynb` file you would like to run &mdash;
+such as `spam_tutorial.ipynb` &mdash; and execute the cells in sequence.
+
+Once an image has been built, you can run it without the `--build` flag:
+
+```bash
+python3 scripts/docker_launch.py spam
+```
+
+### <a name="install"> Installing yourself </a>
+
+Running a tutorial has three required steps if you're installing yourself:
+
 1. Installing repo-wide requirements
 1. Installing tutorial-specific requirements
 1. Launching a Jupyter notebook server or executing as a script
 
 We recommend installing requirements in a virtual environment using [`virtualenv`](https://virtualenv.pypa.io/en/latest/) or [`conda`](https://docs.conda.io/en/latest/).
 For example, if you use pip, first activate your virtualenv if you're using one, then run:
-```
+
+```bash
 # Install requirements (both shared and tutorial-specific)
 pip3 install -r requirements.txt
 pip3 install -r spam/requirements.txt
 
 # Launch the Jupyter notebook interface
 jupyter notebook
 ```
-Then, in the browser tab that opens, navigate to a `.ipynb` file you would like to run, such as `snorkel-tutorials/spam/spam_tutorial.ipynb`, and click to open it. 
-Then execute the cells in sequence.
+Then in the browser tab that opens, navigate to a `.ipynb` file you would like to run &mdash;
+such as `spam/spam_tutorial.ipynb` &mdash; and execute the cells in sequence.
 
 Alternatively, you can run the tutorial as a script by calling `python3` on the corresponding `.py` file directly (e.g. `python3 spam_tutorial.py`).
 

scripts/docker_launch.py

@@ -0,0 +1,62 @@
+import argparse
+import errno
+import os
+import socket
+import subprocess
+
+
+def check_docker() -> None:
+    try:
+        subprocess.run(["docker", "--version"], check=True)
+    except (FileNotFoundError, subprocess.CalledProcessError):
+        raise ValueError("Error calling Docker. Is it installed?")
+
+
+def build_image(tutorial_name: str) -> None:
+    arg = f"TUTORIAL={tutorial_name}"
+    tag = f"--tag={tutorial_name}"
+    subprocess.run(["docker", "build", "--build-arg", arg, tag, "."], check=True)
+
+
+def check_port(port: int) -> None:
+    s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+    try:
+        s.bind(("127.0.0.1", port))
+    except socket.error as e:
+        if e.errno == errno.EADDRINUSE:
+            raise ValueError(f"Port {port} is already in use")
+        else:
+            raise e
+
+
+def run_image(tutorial_name: str, port: int) -> None:
+    tag = f"{tutorial_name}:latest"
+    p_cfg = f"{port}:{port}"
+    p_arg = f"--port={port}"
+    check_run = subprocess.run(["docker", "run", "-it", "-p", p_cfg, tag, p_arg])
+    if check_run.returncode:
+        raise ValueError(
+            "Error running container. If you haven't built it yet, "
+            "try running this script with the --build flag."
+        )
+
+
+def docker_launch(tutorial_name: str, build: bool, port: int) -> None:
+    if not os.path.isdir(tutorial_name):
+        raise ValueError(f"{tutorial_name} is not a valid tutorial")
+    check_docker()
+    if build:
+        build_image(tutorial_name)
+    check_port(port)
+    run_image(tutorial_name, port)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Process some integers.")
+    parser.add_argument("tutorial_name", help="Name of the tutorial (directory)")
+    parser.add_argument(
+        "--build", action="store_true", default=False, help="Build the Docker image?"
+    )
+    parser.add_argument("--port", type=int, default=8888, help="Jupyter port for host")
+    args = parser.parse_args()
+    docker_launch(**vars(args))