amineHY

# Implementation of CNN/ConvNet Model using PyTorch (depicted in the picture above)

class CNN(torch.nn.Module):

    def __init__(self):
        super(CNN, self).__init__()
        # L1 ImgIn shape=(?, 28, 28, 1)
        #    Conv     -> (?, 28, 28, 32)
        #    Pool     -> (?, 14, 14, 32)
        self.layer1 = torch.nn.Sequential(

amineHY

import torch
import torchvision.datasets as dsets


batch_size = 32

# MNIST dataset
mnist_train = dsets.MNIST(root='MNIST_data/',
                          train=True,
                          transform=transforms.ToTensor(),

amineHY

print('Training the Deep Learning network ...')

learning_rate = 0.001
criterion = torch.nn.CrossEntropyLoss()    # Softmax is internally computed.
optimizer = torch.optim.Adam(params=model.parameters(), lr=learning_rate)

train_cost = []
train_accu = []
batch_size = 32
training_epochs = 15

amineHY

def triplet_loss_function(im_anchor, im_positive, im_negative, alpha = 0.2):
    """
    Implementation of the triplet loss function
    
    Source: https://arxiv.org/pdf/1503.03832.pdf
    
    Arguments:
    y_pred -- python list containing three objects:
            anchor -- the encodings for the anchor images, of shape (None, 128)
            positive -- the encodings for the positive images, of shape (None, 128)

amineHY

xhost +

docker run -it --rm -v $(pwd):/workspace \
--runtime=nvidia -w /workspace \
-v /tmp/.X11-unix:/tmp/.X11-unix \
-e DISPLAY=$DISPLAY \
-p 8888:8888 -p 6006:6006 
aminehy/ai-lab:latest

amineHY

Keybase proof

I hereby claim:

• I am aminehy on github.
• I am amine_hy (https://keybase.io/amine_hy) on keybase.
• I have a public key ASCuvs4CY-NYKpzwhW2V2aA9nBkxefzj8bRld7Vvx_TRGAo

To claim this, I am signing this object:

amineHY

name: app

channels:
  - conda-forge
  - anaconda
  
dependencies:
  - flask
  - gunicorn
  - pip:

amineHY

FROM  continuumio/miniconda3
LABEL Author, Amine HadjYoucef

ENV APP_HOME /app
WORKDIR $APP_HOME
COPY . $APP_HOME

#---------------- Prepare the envirennment
RUN conda update --name base conda &&\
    conda env create --file environment.yaml

amineHY

import numpy as np


print('Hello World !')
print('Numpy version is ', np.__version__)

amineHY

.
├── Dockerfile
├── environment.yaml
├── main.py
└── requirements.py