alibi_example

import argparse
import matplotlib.pyplot as plt
from tensorflow.keras.applications.inception_v3 import preprocess_input, decode_predictions
import numpy as np
import requests
import json
import os
from PIL import Image
import ssl
ssl._create_default_https_context = ssl._create_unverified_context


def get_image_data(image_path):
    data = []
    image_shape = (299, 299, 3)
    target_size = image_shape[:2]
    image = Image.open(image_path).convert('RGB')
    image = np.expand_dims(image.resize(target_size), axis=0)
    data.append(image)
    data = np.concatenate(data, axis=0)
    return data


def predict(url, headers, cookies, image_path):
    data = get_image_data(image_path)
    images = preprocess_input(data)

    payload = {
      "instances": [images[0].tolist()]
    }

    # sending post request to TensorFlow Serving server
    print("Calling ", url)
    r = requests.post(url, json=payload, headers=headers, cookies=cookies)
    resp_json = json.loads(r.content.decode('utf-8'))
    preds = np.array(resp_json["predictions"])
    label = decode_predictions(preds, top=1)

    plt.imshow(data[0])
    plt.title(label[0])
    plt.show()


def explain(url, headers, cookies, image_path):
    data = get_image_data(image_path)
    images = preprocess_input(data)

    payload = {
        "instances": [images[0].tolist()]
    }

    # sending post request to TensorFlow Serving server
    print("Calling ", url)
    r = requests.post(url, json=payload, headers=headers, cookies=cookies)
    if r.status_code == 200:
        explanation = json.loads(r.content.decode('utf-8'))

        f, axarr = plt.subplots(1, 2)
        axarr[0].imshow(data[0])
        axarr[1].imshow(explanation['data']['anchor'])
        plt.show()
    else:
        print("Received response code and content", r.status_code, r.content)