jeanmidevacc

import tensorflow as tf
import tensorflow_hub as hub

# colelct the feature extractor of mobile net
feature_extractor_url = "https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/2" #@param {type:"string"}
feature_extractor_layer = hub.KerasLayer(feature_extractor_url,
                                         input_shape=(IMG_HEIGHT, IMG_WIDTH,3))

feature_extractor_layer.trainable = False

jeanmidevacc

import tensorflow as tf

model = tf.keras.Sequential([
    tf.keras.layers.Conv2D(
        32, (3, 3), padding='same', activation='relu'),
    tf.keras.layers.MaxPooling2D((2, 2)),
    tf.keras.layers.Dropout(rate=0.5),
    
    tf.keras.layers.Conv2D(
        64, (3, 3), padding='same', activation='relu'),

jeanmidevacc

import tensorflow as tf

model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(input_shape=(IMG_HEIGHT, IMG_WIDTH, 3)),
    tf.keras.layers.Dense(300, activation="relu"),
    tf.keras.layers.Dense(100, activation="relu"),
    tf.keras.layers.Dense(len(CLASS_NAMES), activation="softmax")
])

model.compile(optimizer='adam',

jeanmidevacc

import tensorflow as tf

model = tf.keras.models.Sequential([
tf.keras.layers.Conv2D(16, 3, padding='same', activation='relu', input_shape=(IMG_HEIGHT, IMG_WIDTH ,3)),
tf.keras.layers.MaxPooling2D(),
tf.keras.layers.Conv2D(32, 3, padding='same', activation='relu'),
tf.keras.layers.MaxPooling2D(),
tf.keras.layers.Conv2D(64, 3, padding='same', activation='relu'),
tf.keras.layers.MaxPooling2D(),
tf.keras.layers.Flatten(),

jeanmidevacc

"""
Based on https://www.tensorflow.org/tutorials/images/cnn
"""

import pathlib
import tensorflow as tf

# Definition of the constant
BATCH_SIZE = 30
EPOCHS = 5

jeanmidevacc

informations = []
for i,run in enumerate(runs):
    
    if run.successful:
        # collect some details on the fisrt and last step of the flow
        step_start = Step(f"{flowname}/{run.id}/start")
        step_end = run.end_task
        
        # Collect the number of cards picked for the features computation
        nbr_cardsselected = step_start.task.data.limittopcards

jeanmidevacc

"""
pipeline.py
Script to test the different decorator on the metaflow framework
"""

import random
from metaflow import FlowSpec, step, Parameter, conda, conda_base

@conda_base(disabled = False ,python="3.7.4", libraries={"pandas" : "0.25.2"})
class ExampleFlow(FlowSpec):

jeanmidevacc

from PIL import Image

def build_mlimage(path, config_resize = (100,50), is_bw = True):
    # Access the image
    img = Image.open(path)
    
    # Resizing and conversion in black and white (if necessary)
    if is_bw:
        newimg = img.resize(config_resize, Image.ANTIALIAS).convert('L')
    else:

jeanmidevacc

from sklearn.neighbors import KNeighborsRegressor
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score, explained_variance_score

import mlflow
import mlflow.sklearn

import numpy as np

# Launch the experiment on mlflow
experiment_name = "electricityconsumption-forecast"

jeanmidevacc

import boto3
import json

# Name of the app that you defined during the deployment on sagemaker
app_name = "xxxxx"
# AWS region of the deployment of the app on sagemaker
region = "xxxxx"

# Function to collect data from the endpoint on sagemaker
def query_endpoint(input_json):