Tensorflow low-level API implementations for various Machine Learning algorithms
Models were implemented using Tensorflow 1.3. Create a conda environment using the .yml file in this repository
-
nn_mnist_tf.ipynb:
- Fully connected deep neural network model using MNIST dataset.
-
mnist_cnn.ipynb
- Convolution neural network model using MNIST dataset.
- 2 convolutional + max pooling layers followed by two fully connected layers
-
time_series_predictive_model_rnn.ipynb
- Recurrent neural network to model some time series data using the monthly-milk-production.csv dataset.
-
auto_encoder.ipynb
- Linear autoencoder able to separate a 30 dimensional dataset using only two neurons