Prácticas de Pytorch

Pytorch es una librería reciente que se está esparciendo en la comunidad académina e industrial para desarrollar sistemas basados en aprendizaje automático, en especial en aprendizaje profundo. En este repositorio encontrarás una serie de ejercicios para aprender o reforzar tu conocimiento acerca de la implementación de redes neuronales profundas. Este repositorio es complemento al curso de introducción a las redes neuronales. Para más detalles visita mi página web.

Irving Vasquez Sitio Web

Configurar ambiente con conda

Para facilitar la instalación y ejecución de los ejercicios se utiliza de manejador conda. En dependencia de que distribución se use se puede llamar minicomda o anaconda. Para simplicidad en este repositorio se usa minconda.

De la definición de conda docs:

Conda is an open source package management system and environment management system for installing multiple versions of software packages and their dependencies and switching easily between them. It works on Linux, OS X and Windows, and was created for Python programs but can package and distribute any software.

Pasos

Para tener los programas listos haremos dos cosas:

1. Install miniconda on your computer, by selecting the latest Python version for your operating system. If you already have conda or miniconda installed, you should be able to skip this step and move on to step 2.
2. Create and activate * a new conda environment.

* Nota que cada que vayamos a usar los programas debemos de activar el ambiente de conda!

---

1. Instalación de miniconda

Download the latest version of miniconda that matches your system.

NOTE: There have been reports of issues creating an environment using miniconda v4.3.13. If it gives you issues try versions 4.3.11 or 4.2.12 from here.

	Linux	Mac	Windows
64-bit	64-bit (bash installer)	64-bit (bash installer)	64-bit (exe installer)
32-bit	32-bit (bash installer)		32-bit (exe installer)

Install miniconda on your machine. Detailed instructions:

• Linux: http://conda.pydata.org/docs/install/quick.html#linux-miniconda-install
• Mac: http://conda.pydata.org/docs/install/quick.html#os-x-miniconda-install
• Windows: http://conda.pydata.org/docs/install/quick.html#windows-miniconda-install

2. Crear y activar el ambiente

For Windows users, these following commands need to be executed from the Anaconda prompt as opposed to a Windows terminal window. For Mac, a normal terminal window will work.

Git and version control

These instructions also assume you have git installed for working with Github from a terminal window, but if you do not, you can download that first with the command:

conda install git

Now, we're ready to create our local environment!

1. Clone the repository, and navigate to the downloaded folder. This may take a minute or two to clone due to the included image data.

git clone https://github.com/irvingvasquez/practicas_pytorch.git
cd practicas_pytorch

2. Create (and activate) a new environment, named pptorch with Python 3.7. If prompted to proceed with the install (Proceed [y]/n) type y.

   • Linux or Mac:

   conda create -n pptorch python=3.7
   source activate pptorch
   

   • Windows:

   conda create --name pptorch python=3.7
   conda activate pptorch
   

   At this point your command line should look something like:

   (pptorch) <user>:practicas_pytorch <user>$.

   The (pptorch) indicates that your environment has been activated, and you can proceed with further package installations.

3. Install PyTorch and torchvision; this should install the latest version of PyTorch. Mi recomendación es revisar antes la documentación oficial de pytorch y verificar los comandos en dependencia de si se va a utilizar GPU o no. Los siguientes comandos son para usar CPU.

   • Linux or Mac:

   conda install pytorch=1 torchvision cpuonly -c pytorch
   

   • Windows:

   conda install pytorch=1 torchvision cpuonly -c pytorch
   

4. Install a few required pip packages, which are specified in the requirements text file (including OpenCV).

pip install -r requirements.txt

7. That's it!

Now all of the pptorch libraries are available to you. Assuming you're environment is still activated, you can navigate to the Exercises repo and start looking at the notebooks:

cd
cd practicas_pytorch
jupyter notebook

To exit the environment when you have completed your work session, simply close the terminal window.

Referencias usadas: Udacity, Deep Learning with PyTorch