Merge pull request #2 from MGTheTrain/feature/ml-training-app

[Feature] Simple feedforward neural network with MNIST dataset to map input images to their corresponding digit classes

CHANGELOG.md

@@ -15,4 +15,5 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [Feature] kubeflow operator or mlflow helm chart installations in deployed AKS clusters
 - [Feature] CD workflow for on-demand AKS deployments and kubeflow operator or mlflow helm chart installations
 - [Feature] CD wofklow for on demand deployments of an Azure Storage Account Container **(For storing terraform state files)**
-- [Feature] Added `devcontainer.json` with necessary tooling for local development
+- [Feature] Added `devcontainer.json` with necessary tooling for local development
+- [Feature] Simple feedforward neural network with MNIST dataset to map input images to their corresponding digit classes 

README.md

@@ -24,10 +24,11 @@ Repository showcasing ML Ops practices with kubeflow and mlflow
 - [x] CD workflow for on-demand AKS deployments and kubeflow operator or mlflow helm chart installations
 - [x] CD wofklow for on demand deployments of an Azure Storage Account Container **(For storing terraform state files)**
 - [x] Added `devcontainer.json` with necessary tooling for local development
-- [ ] Dockerized Python (pytorch or tensorflow) application for ML training purposes or Jupyter notebooks
-    - [ ] CNN architecture training considering pre-trained models for image classification AI applications
+- [ ] Dockerized Python (pytorch or tensorflow) application for ML training purposes and Jupyter notebooks
+    - [x] Simple feedforward neural network with MNIST dataset to map input images to their corresponding digit classes 
+    - [ ] CNN architecture training considering COCO dataset and pre-trained models for image classification AI applications
     - [ ] (**OPTIONAL**) Transformer architecture training considering pre-trained models for chatbot AI applications
-- [ ] Helm charts with K8s manifests for ML jobs (with and without kubeflow)
+- [ ] Helm charts with K8s manifests for ML jobs (with and without kubeflow) considering the [Training Operator for CRDs](https://github.com/kubeflow/training-operator)
 - [ ] Demonstration of model training and model deployment trough automation workflows
 - [ ] (**OPTIONAL**) mlflow experiments for the machine learning lifecycle
 
@@ -77,6 +78,14 @@ The volumes that were created appear as follows:
 
 ![Jupyter instance created volumes](./images/jupyter-instance-created-volumes.PNG)
 
+The Jypter instace that was created appear as follows:
+
+![Created Jupyter instance](./images/created-jupyter-instance.PNG)
+
+Once `CONNECTED` to a Jupyter instance ensure to clone this Git repository (HTTPS URL: `https://github.com/MGTheTrain/ml-ops-ftw.git`):
+
+![Clone git repository](./images/clone-git-repository.PNG)
+
 TBD
 
 

python/README.md → python/keras-mnist-training/README.md

@@ -1,4 +1,4 @@
-# python
+# Keras training app with mnist dataset
 
 ### Folder structure
 

python/keras-mnist-training/main.py

@@ -0,0 +1,4 @@
+from src import train
+
+if __name__ == "__main__":
+    train.main()

python/keras-mnist-training/requirements.txt

@@ -0,0 +1,2 @@
+tensorflow=2.16.1
+numpy

python/keras-mnist-training/src/data_loader.py

@@ -0,0 +1,14 @@
+import tensorflow as tf
+
+def load_data():
+    # Load MNIST dataset
+    (x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()
+
+    # Normalize pixel values to the range [0, 1]
+    x_train, x_test = x_train / 255.0, x_test / 255.0
+
+    # Add channel dimension for convolutional networks
+    x_train = x_train[..., tf.newaxis]
+    x_test = x_test[..., tf.newaxis]
+
+    return (x_train, y_train), (x_test, y_test)

python/keras-mnist-training/src/model.py

@@ -0,0 +1,9 @@
+import tensorflow as tf
+from tensorflow.keras import layers, models
+
+def build_model():
+    model = models.Sequential([
+        layers.Dense(64, activation='relu', input_shape=(28, 28)),
+        layers.Dense(10, activation='softmax')
+    ])
+    return model

python/keras-mnist-training/src/train.py

@@ -0,0 +1,11 @@
+from .data_loader import load_data
+from .model import build_model
+
+def main():
+    x_train, y_train = load_data()
+    model = build_model()
+    model.compile(optimizer='adam',
+                  loss='sparse_categorical_crossentropy',
+                  metrics=['accuracy'])
+
+    model.fit(x_train, y_train, epochs=10, batch_size=32)

python/keras-mnist-training/src/utils.py

@@ -0,0 +1,3 @@
+def preprocess_data(data):
+    # Data preprocessing code here
+    pass

python/tf-yolo-coco-training/.gitignore

@@ -0,0 +1,39 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+env/
+build/
+dist/
+eggs/
+*.egg-info/
+
+# IDEs
+.idea/
+.vscode/
+
+# Dependency directories
+venv/
+env/
+
+# Compiled Python files
+*.pyc
+*.pyo
+
+# Logs and databases
+*.log
+*.sqlite
+*.db
+
+# Model artifacts
+*.pt  # PyTorch model files
+*.h5  # TensorFlow model files
+
+# Miscellaneous
+.DS_Store

python/tf-yolo-coco-training/README.md

@@ -0,0 +1,25 @@
+# Tensorflow yolov4 training app with coco dataset
+
+### Folder structure
+
+```sh
+project_root/
+│
+├── data/                   # Directory for storing datasets
+│   └── dataset/
+│       └── ...
+│
+├── models/                 # Directory for storing trained models
+│   └── saved_models/
+│       └── ...
+│
+├── src/                    # Source code directory
+│   ├── data_loader.py      # Data loading utilities. Load the dataset from the specified directory (data/ in this case) or from external sources like databases or APIs
+│   ├── model.py            # Model architecture definition
+│   ├── train.py            # Training script
+│   └── utils.py            # Utility functions
+│
+├── requirements.txt        # Python dependencies
+├── README.md               # Project README file
+└── main.py                 # Main entry point script for running the application
+```

python/tf-yolo-coco-training/main.py

@@ -0,0 +1,4 @@
+from src import train
+
+if __name__ == "__main__":
+    train.main()

python/tf-yolo-coco-training/requirements.txt

@@ -0,0 +1,2 @@
+tensorflow=2.16.1
+numpy

python/tf-yolo-coco-training/src/data_loader.py

@@ -0,0 +1,4 @@
+import tensorflow as tf
+
+def load_data():
+    pass

python/tf-yolo-coco-training/src/model.py

@@ -0,0 +1,3 @@
+
+def build_model():
+    pass

python/tf-yolo-coco-training/src/train.py

@@ -0,0 +1,5 @@
+from .data_loader import load_data
+from .model import build_model
+
+def main():
+    pass

python/tf-yolo-coco-training/src/utils.py

@@ -0,0 +1,3 @@
+def preprocess_data(data):
+    # Data preprocessing code here
+    pass