rocks-classifier 0.0.4

Rock classifier deployed on railway and monitored using Weights and Biases!

Whats-this-rock

This project deploys a telegram bot that classifies rock images into 1 of 7 types.

Installation & Training Steps

Use the Telegram Bot

You can try the bot here on Telegram.

Type /help to get instructions.

Deploy Telegram Bot

pip install -r requirements-prod.txt
python rock_classifier/bot.py

Train Model

Paste your kaggle.json file in the root directory

Run these commands

pip install -r requirements-dev.txt
sh src/scripts/setup.sh
python src/models/train.py

You can try different models and parameters by editing config.json.

By using Hydra it’s now much more easier to override parameters like this

python src/models/train.py  wandb.project=Whats-this-rockv \
                            dataset_id=[1,2,3,4] \
                            epochs=50 \
                            backbone=resnet

Wandb Sweeps (Hyperparameter Tuning)

Edit configs/sweeps.yaml

wandb sweep \
--project Whats-this-rock \
--entity udaylunawat \
configs/sweep.yaml

This will return a command with $sweepid

wandb agent udaylunawat/Whats-this-rock/$sweepid

Demo

Run in Colab	View Source on GitHub	Download Notebook

Features

\Features added

\Features planned

Wandb Datasets 4 Datasets Augmentation keras-cv Regular Augmentation Sampling Oversampling Undersampling Class weights Remove Corrupted Images Try Multiple Optimizers (Adam, RMSProp, AdamW, SGD) Generators TFDS datasets ImageDataGenerator Models ConvNextTiny BaselineCNN Efficientnet Resnet101 MobileNetv1 MobileNetv2 Xception LRScheduleer, LRDecay Baseline without scheduler Step decay Cosine annealing Classic cosine annealing with bathc steps w/o restart Model Checkpoint, Resume Training Evaluation Confusion Matrix Classification Report Deploy Telegram Bot Heroku - Deprecated Railway Show CM and CL in bot Docker GitHub Actions Deploy Bot when bot.py is updated. Lint code using GitHub super-linter Configuration Management ml-collections Hydra Performance improvement Convert to tf.data.Dataset Linting & Formatting Black Flake8 isort pydocstyle

Deploy to Huggingface spaces Accessing the model through FastAPI (Backend) Streamlit (Frontend) convert models.py to Classes and more OOP style nbdev Group Runs kfold cross validation WandB Tables find the long tail examples or hard examples, find the classes that the model is performing terribly on, Add Badges Linting Railway

Technologies Used

Directory Tree

├── imgs                              <- Images for skill banner, project banner and other images
│
├── configs                           <- Configuration files
│   ├── configs.yaml                  <- config for single run
│   └── sweeps.yaml                   <- confguration file for sweeps hyperparameter tuning
│
├── data
│   ├── corrupted_images              <- corrupted images will be moved to this directory
│   ├── sample_images                 <- Sample images for inference
│   ├── 0_raw                         <- The original, immutable data dump.
│   ├── 1_external                    <- Data from third party sources.
│   ├── 2_interim                     <- Intermediate data that has been transformed.
│   └── 3_processed                   <- The final, canonical data sets for modeling.
│
├── notebooks                         <- Jupyter notebooks. Naming convention is a number (for ordering),
│                                        the creator's initials, and a short `-` delimited description, e.g.
│                                        1.0-jqp-initial-data-exploration`.
│
│
├── src                               <- Source code for use in this project.
│   │
│   ├── data                          <- Scripts to download or generate data
│   │   ├── download.py
│   │   ├── preprocess.py
│   │   └── utils.py
│   │
│   ├── callbacks                     <- functions that are executed during training at given stages of the training procedure
│   │   ├── custom_callbacks.py
│   │   └── callbacks.py
│   │
│   ├── models                        <- Scripts to train models and then use trained models to make
│   │   │                                predictions
│   │   ├── evaluate.py
│   │   ├── models.py
│   │   ├── predict.py
│   │   ├── train.py
│   │   └── utils.py
│   │
│   └── scripts                       <- Scripts to setup dir structure and download datasets
│   │   ├── clean_dir.sh
│   │   ├── dataset1.sh
│   │   ├── dataset2.sh
│   │   ├── dataset3.sh
│   │   ├── dataset4.sh
│   │   └── setup.sh
│.  │
│   └── visualization                 <- Scripts for visualizations
│
├── .dockerignore                     <- Docker ignore
├── .gitignore                        <- GitHub's excellent Python .gitignore customized for this project
├── LICENSE                           <- Your project's license.
├── Makefile                          <- Makefile with commands like `make data` or `make train`
├── README.md                         <- The top-level README for developers using this project.
├── requirements.txt                  <- The requirements file for reproducing the analysis environment, e.g.
│                                        generated with `pip freeze > requirements.txt`
└── setup.py                          <- makes project pip installable (pip install -e .) so src can be imported

Bug / Feature Request

If you find a bug (the site couldn’t handle the query and / or gave undesired results), kindly open an issue here by including your search query and the expected result.

If you’d like to request a new function, feel free to do so by opening an issue here. Please include sample queries and their corresponding results.

Contributing

• Contributions make the open source community such an amazing place to learn, inspire, and create.
• Any contributions you make are greatly appreciated.
• Check out our contribution guidelines for more information.

License

LinkFree is licensed under the MIT License - see the LICENSE file for details.

Credits

• Dataset - by Mahmoud Alforawi

Support

This project needs a ⭐️ from you. Don’t forget to leave a star ⭐️

Walt might be the one who knocks
but Hank is the one who rocks.