Cats-vs-Dogs

Follow along the jupyter notebook to learn basic programming in tensorflow and tesorboard.

Downloading Dataset

https://www.microsoft.com/en-ca/download/details.aspx?id=54765

About Model

Setup

Clone the repository

  git clone https://github.com/GopiKishan14/Cats-vs-Dogs.git

Opening notebook on local host:

  cd Cats-vs-Dogs
  jupyter-notebook

Futher comments are given in notebooks to help out.

Pre-processing Dataset

Open Dataset-Builder.ipynb

Training Model :-

Open model.ipynb

Performance and Evaluation

Models trained on partial datset and the partial pre-processed datset
is provided.

Saved Model

Built Dataset

Running Tensorboard :-

Check versions :-

from tensorboard import version; 
print(version.VERSION)
import tensorflow as tf; 
print(tf.__version__)

Running log_dir in tensorboard :-

tensorboard --logdir = logs/

[Batch_acc]

[Batch loss]

Predicting :-

Open predict.ipynb

Installing Dependencies :

[To install tensorflow] Refer to this link

Or , on Linux for CPU-only (no GPU)

  pip install -U pip
  pip install tensorflow

To install anaconda-navigator

conda install -c anaconda anaconda-navigator

To install tensorboard

conda install -c conda-forge tensorboard

Or through pip :-

pip install tensorboard pip show tensorboard

To install opencv:-

  pip install opencv-python
  python 3.x
  Numpy
  scipy
  matplotlib

About Dataset

Asirra (Animal Species Image Recognition for Restricting Access) is a HIP that works by asking users to identify photographs of cats and dogs. This task is difficult for computers, but studies have shown that people can accomplish it quickly and accurately. Asirra is unique because of its partnership with Petfinder.com, the world’s largest site devoted to finding homes for homeless pets. They’ve provided Microsoft Research with over three million images of cats and dogs, manually classified by people at thousands of animal shelters across the United States. Kaggle is fortunate to offer a subset of this data for fun and research. For more information, see https://www.kaggle.com/c/dogs-vs-cats.

Contributing

If you wish to contribute, feel free to open a PR Peace out!