This file is used to send http request to endpoint created using Tensorflow Serving.

request.py

import os
from os import path
import json
import requests
import numpy as np
import matplotlib.pyplot as plt
import subprocess
import gzip
import sys

os.chdir(os.path.dirname(os.path.abspath(__file__)))

class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',
               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']

# Download Test mnist fashion dataset
if not path.exists("data/t10k-images-idx3-ubyte.gz") or not path.exists("data/t10k-labels-idx1-ubyte.gz"):
  subprocess.run(["wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz -P data/"],shell=True)
  subprocess.run(["wget http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz -P data/"],shell=True)

# Reading Test Images
f = gzip.open('data/t10k-images-idx3-ubyte.gz','r')
image_size = 28
num_images = 3
f.read(16)
buf = f.read(image_size * image_size * num_images)
img_data = np.frombuffer(buf, dtype=np.uint8).astype(np.float32)
img_data = img_data.reshape(num_images, image_size, image_size, 1)

# Reading Test labels
f = gzip.open('data/t10k-labels-idx1-ubyte.gz','r')
f.read(8)
test_labels = []
for i in range(0,3):   
    buf = f.read(1)
    test_labels.append(np.frombuffer(buf, dtype=np.uint8).astype(np.int64)[0])

def show(idx,title):
    image = np.asarray(img_data[idx]).squeeze()
    plt.imshow(image)
    plt.axis('off')
    plt.title('\n{}'.format(title), fontdict={'size': 9})
    plt.show()

with open('/path to the json file/predict.json') as f:
  data = json.load(f)
json_data = json.dumps(data)

headers = {"content-type": "application/json"}
json_response = requests.post('http://localhost:8501/v1/models/fashion_mnist:predict', data=json_data, headers=headers)

predictions = json.loads(json_response.text)['predictions']
print(predictions)

for i in range(0,3):
    show(i, 'The model thought this was a {} (class {}), and it was actually a {} (class {})'.format(
  class_names[np.argmax(predictions[i])], np.argmax(predictions[i]), class_names[test_labels[i]], test_labels[i]))