shopty

shopty is a tool for tuning hyperparameters on your computer or slurm-managed clusters.


Keywords
deep-learning, machine-learning, python, pytorch
License
MIT
Install
pip install shopty==0.0.3

Documentation

shopty

Generic badge Code style: black

Simple Hyperparameter OPTimization in pYthon

Install from source (recommended)

git clone https://github.com/colligant/shopty
# optional: pip install flit
cd shopty && flit install

Install via pip

pip install shopty

Run the examples with commands here. A non-cli example is here.

What is the purpose of this tool?

Lots of other hyperparameter tuning libraries (at least the ones I've found, anyways) require modifying a bunch of source code and make assumptions about your deployment environment.

shopty is a simple library to tune hyperparameters either on your personal computer or a slurm-managed cluster that requires minimal code changes and uses a simple config file to do hyperparameter sweeps.

Design

The Supervisor classes in shopty spawn (if on CPU) or submit (if on slurm) different experiments, each with their own set of hyperparameters. Submissions are done within python by creating a bash or sbatch file and submitting it via subprocess.call.

Each experiment writes a "results.txt" after its finished to a unique directory. The Supervisor class detects when each experiment is done and reads the "results.txt" file for the outcome of the experiment that wrote it.

Source code modifications

See a simple example here. A neural network example is here.

Supervisors communicate with experiments via environment variables. Your custom training code must know how to deal with some shopty-specific use cases. In particular, it must a) run the code for max_iter iterations, b) reload the training state from a checkpoint file, and c) write the result post-training to a results file. The checkpoint filepath, results filepath, and maximum iteration to run for are all provided by shopty. The python files in the examples/ directory show how to achieve (a, b, and c) with a simple non-nn example and a Pytorch Lightning neural network example.

How to define hyperparameters and slurm directives

We use a .yaml file to define hyperparameters for training models as well as other commands you want to run to set up the training environment. See after the yaml markup for header-specific information. An example .yaml file:

project_name: 'your_project_name'
run_command: "python3 my_cool_script.py"
project_dir: "~/deep_thought/"
monitor: "max"
poll_interval: 10

hparams:
  learning_rate:
    begin: -10
    end: -1
    random: True
    log: True
  your_custom_hparam:
    begin: 1
    end: 5
    step: 1 
  another_custom_hparam:
    begin: 1
    end: 5
    random: True
  
statics:
  a_static_hparam: 1e-10

slurm_directives:
  - "--partition=gpu"
  - "--gres=gpu:1"

environment_commands:
  - "conda activate my_env"

By default all hyperparameters will be interpreted as floats. Add a type: field in the hyperparameter definition to specify types.

hparam_2:
   begin: 100
   end: 300
   step: 40
   type: 'int' # or 'float'

run_command

The run_command is how shopty runs your program. Generated hyperparameters are passed in to the run_command via the command line in no particular order. For example, if you want to tune the learning rate of the model in my_cool_script.py, my_cool_script.py must accept a --learning_rate argument.

project_name, project_dir, monitor, and poll_interval

project name is what your experiments will be titled. project_dir is where output and logs will be saved. You can minimize or maximize metrics with the monitor field - 'min' and 'max' are supported. poll_interval is how often shopty polls processes for completion in seconds.

hparams

The hparams header has two levels of indentation: one for the name of hyperparameter, and the next for the beginning and end of the range over which to sample from. There are three required elements for each hparam: begin, end, and <random or step>. The hyperparameter can either be sampled randomly between the interval [begin, end) or iterated over from begin to end with step step. Binary variables can be added to the project with

hparams:
  binary_indicator:
    begin: 0
    end: 2
    step: 1

shopty automatically assumes hyperparameters are floats, but you can add a type with

hparams:
  my_int_hparam:
    begin: -10
    end: 10
    step: 1
    type: 'int'

Options: 'float' 'int'. String hyperparameters are not supported.

statics

Static variables can be added under the static header:

statics:
    my_static_var: 10
    # or, if you need to specify a type:
    my_other_static_var:
        val: 100.0
        type: 'float'

Slurm directives

Slurm scripts have headers that specify what resources a program will use (#SBATCH statements). Add these to each experiment by editing the slurm_directives section of the yaml file. They will be added as #SBATCH statements in each slurm submission script.

Environment commands

These are arbitrary commands that you want to run before the run_command is called in the generated script.

Shopty isn't working on slurm. How do I debug this?

shopty generates slurm scripts to run each experiment. Navigate to an experiment's directory, request an interactive node, and run the slurm script with bash slurm_script.sh.