Task 2

The work has been done using Google Colab.

https://colab.research.google.com/

Each part of the task is done separately, Point 1 as subtask 1 and likewise.
Model weights and architecture either used or created are in there respective subtasks directory.

| --- Subtask1-> Subtask_1.ipynb
| --- Subtask2-> Subtask_2.ipynb
| --- Subtask3-> Subtask_3.ipynb

Installation

The requirement.txt file has been exported using google colab.
Because of which list might seem to long but it's very quick to install in colab.

pip install -r requirements.txt

Reference

https://www.tensorflow.org/api_docs/python/tf

https://www.machinecurve.com/index.php/2019/09/16/he-xavier-initialization-activation-functions-choose-wisely/

https://heartbeat.fritz.ai/pre-trained-machine-learning-models-vs-models-trained-from-scratch-63e079ed648f

https://www.baeldung.com/cs/ml-neural-network-weights

https://thispointer.com/how-to-change-current-working-directory-in-python/

https://developers.google.com/machine-learning/data-prep/construct/collect/data-size-quality

https://stackoverflow.com/questions/54057011/google-colab-session-timeout