Code-Snippets, tutorials and librarys for AI algorithm development
If you do not want to setup the environment on your computer, you can also execute your jupyter-notebooks on:
- Google Colab https://colab.research.google.com/drive/
- Kaggle https://www.kaggle.com/
- AWS https://aws.amazon.com/
git clone https://github.com/maximilianharr/plai plai
cd docker
docker-compose build
docker-compose up
Latest versions of pandas and scikit-learn prevent import of pandas_ml v.0.6.1. To fix this, install supported versions
python3 -m pip install pandas==0.24.0
python3 -m pip install scikit-learn==0.20.0
Run all unittests using
python3 ${PATH_TO_WS}/common/tools/run_all_unittests.py
./common/tutorials/README.md https://docs.google.com/presentation/d/1TeR-_v_WoZFvz8AkOEy-XYhMKhS0AVy31fzfF2hcAQQ/edit?usp=sharing
- "Machine Learning von A-Z: Lerne Python & R für Data Science!" by Jannis Seemann
https://www.udemy.com/course/machine-learning-komplett/ - "Deep Learning, Neuronale Netze & AI: Der Komplettkurs" by Jannis Seemann
https://www.udemy.com/course/deep-learning-und-ai/ - "Complete Guide to TensorFlow for Deep Learning with Python" by Jose Portilla"
https://www.udemy.com/course/complete-guide-to-tensorflow-for-deep-learning-with-python/
https://pypi.org/project/darknetpy/ https://pypi.org/project/yolov3/
https://neurohive.io/en/
Implementations of common nets: https://github.com/keras-team/keras-applications
Trained common nets: https://keras.io/api/applications/
Some popular datasets are listed at https://www.tensorflow.org/datasets/catalog/overview
Coco (Common Objects in Context) http://cocodataset.org/#home
Openimages https://storage.googleapis.com/openimages/web/index.html
jupyter-notebook
jupyter nbconvert --to=python --output-dir=./ ${SCRIPT}.ipynb
pip install jupyterthemes
jt -t chesterish
https://github.com/facebookresearch/detr
Problem:
pandas_ml v.0.6.1 supports pandas v0.22.0 and scikit-learn 0.20.0 (1) and will throw "AttributeError" if newer version of pandas and scikit-learn provided.
(1) https://pandas-ml.readthedocs.io/en/latest/
Solution (not tested):
python3 -m pip install pandas==0.24.0
python3 -m pip install scikit-learn==0.20.0
sudo apt install rustc
pip install --upgrade pip setuptools wheel
python3 -m pip install bottleneck
python3 -m pip install darknetpy
python3 -m pip install yolov3