omarsar

GET blogs_fixed/_search
{
    "query": {
        "query_string" : {
            "query" : "(content:elasticsearch AND content:engineering) OR (title:elasticsearch AND title:engineering)"
        }
    }
}

omarsar

GET _search
{
  "query": {
    "match_all": {}
  }
}


GET /

omarsar

!pip3 install torch torchvision

omarsar

## our data in tensor form
x = torch.tensor([[-1.0],  [0.0], [1.0], [2.0], [3.0], [4.0]], dtype=torch.float)
y = torch.tensor([[-3.0], [-1.0], [1.0], [3.0], [5.0], [7.0]], dtype=torch.float)

omarsar

## Neural network with 1 hidden layer
layer1 = nn.Linear(1,1, bias=False)
model = nn.Sequential(layer1)

omarsar

## loss function
criterion = nn.MSELoss()

## optimizer algorithm
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

omarsar

## training
for i in range(150):
    model = model.train()

    ## forward
    output = model(x)
    loss = criterion(output, y)
    optimizer.zero_grad()

    ## backward + update model params 

omarsar

## test the model
sample = torch.tensor([10.0], dtype=torch.float)
predicted = model(sample)
print(predicted.detach().item())

omarsar

## Import the usual libraries
import torch
import torchvision
import torch.nn as nn
from torchvision import datasets, models, transforms
import os
import numpy as np
import matplotlib.pyplot as plt

%matplotlib inline

omarsar

## configuration to detect cuda or cpu
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print (device)