🔥 FlameAI

Python Deep Learning Toolkit.

Installation

Install the package:

pip install flameai

Update the package:

python3 -m pip install --upgrade pip
pip3 install --upgrade flameai

Example

Evaluate the performance of a binary classification model:

# simple.py
import flameai

y_true = [0, 0, 0, 1, 0, 1, 0, 1, 1, 0]
y_pred = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]

flameai.eval_binary(y_true, y_pred, threshold=0.5)

$ python examples/simple.py
threshold: 0.50000
accuracy: 0.70000
precision: 0.60000
recall: 0.75000
f1_score: 0.66667
auc: 0.70833
cross-entropy loss: 4.03816
True Positive (TP): 3
True Negative (TN): 4
False Positive (FP): 2
False Negative (FN): 1
confusion matrix:
[[4 2]
 [1 3]]

More examples: examples

Test Locally

Create a conda environment:

# Create env
mamba create -n python_3_10 python=3.10

# Activate env
conda activate python_3_10

# Check envs
conda info --envs

# Deactivate env
conda deactivate

# Remove env
conda env remove --name python_3_10

Install the package from source (or local wheel):

# Check if flameai has been installed
pip list | grep flameai

# Install from source
pip install -e .

# Or install from local wheel
# `pip install dist/flameai-[VERSION]-py3-none-any.whl`

# Uninstall
pip uninstall flameai

# Reinstall
pip uninstall flameai -y && pip install -e .

Test:

# Install pytest
pip install pytest

# Run tests
pytest

# Install nox
pip install nox

# Run nox
nox

Lint:

# Install flake8 and flake8-import-order
pip install flake8
pip install flake8-import-order

# Lint
flake8 --import-order-style google

Development

Build:

python3 -m pip install --upgrade build

python3 -m build

Upload:

python3 -m pip install --upgrade twine

twine upload dist/*