SlideShare uma empresa Scribd logo
1 de 36
BASEL BERN BRUGG DÜSSELDORF FRANKFURT A.M. FREIBURG I.BR. GENEVA
HAMBURG COPENHAGEN LAUSANNE MUNICH STUTTGART VIENNA ZURICH
Machine Learning …
… with Neural Networks, TensorFlow and Keras
Christoph Pletz
Agenda
Name Presentation2 9/26/2018
1. Machine Learning Basics
2. Neural Networks
3. Using TensorFlow and Keras
Machine Learning with Neural Networks, TensorFlow and Keras3 9/26/2018
Machine Learning Basics
Types of Machine Learning
Machine Learning with Neural Networks, TensorFlow and Keras4 9/26/2018
Regression
– How much? How many?
Classification
– Which class does this belong to?
Clustering
– Are there different groups? To which group does an item belong?
Anomaly Detection
– Is this weird?
Recommendation
– Which option to choose?
supervisedunsupervised
Simplest Example – housing prices
Machine Learning with Neural Networks, TensorFlow and Keras5 9/26/2018
Living Space
Price
Living
Space
Price
95 1000000
100 950000
120 1200000
… …
250 3300000
Given a house with a certain living space,
what is the expected price?
Terminology: Features and Labels
Machine Learning with Neural Networks, TensorFlow and Keras6 9/26/2018
Living
Space
Price
95 1000000
100 950000
120 1200000
… …
250 3300000
Feature(s)
x
Label(s)
y
Hypothesis Function / Model
Machine Learning with Neural Networks, TensorFlow and Keras7 9/26/2018
𝑦 = ℎ(𝑥)
hypothesis
model
𝑥 predict
𝑦
𝑦
𝑦 Actual label value
Predicted value
𝑥 𝑦 𝑦
95 1000000 998734
100 950000 1043576
120 1200000 1236556
… …
Second Simplest Example – Iris classification
Machine Learning with Neural Networks, TensorFlow and Keras8 9/26/2018
CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=170298
Iris setosa Iris versicolor Iris virginica
Data collected by Edgar Anderson and published
by Ronald Fisher in 1936 (150 samples)
Iris Data Set
Machine Learning with Neural Networks, TensorFlow and Keras9 9/26/2018
https://upload.wikimedia.org/wikipedia/commons/thumb/5/56/Iris_dataset_scatterplot.svg/749px-Iris_dataset_scatterplot.svg.png
Iris Data Set (classification)
Machine Learning with Neural Networks, TensorFlow and Keras10 9/26/2018
Sepal
length
Sepal
width
Petal
length
Petal
width
Species
5.9 3.0 4.2 1.5 versicolor
6.9 3.1 5.4 2.1 virginica
5.1 3.3 1.7 0.5 setosa
… … … … …
Species
code
1
2
0
…
Is
Setosa
Is
versicolor
Is
virginica
0 1 0
0 0 1
1 0 0
… … …
One hot encoding
Features
x
Labels
y
A bit more challenging: Image Recognition
Machine Learning with Neural Networks, TensorFlow and Keras11 9/26/2018
Lets assume we want to classify the iris species from images …
E.g. 512 * 512 pixels and 3 color channels  786432 features, still 3 classes
Imagenet (http://www.image-net.org/) provides pictures for 1000 classes
model
predict 𝑦
A Linear Model for Housing Prices
Machine Learning with Neural Networks, TensorFlow and Keras12 9/26/2018
x
y
Linear Model
Parameters
(b = 100000, w = 0.5)
𝑦 = 𝑤 ∗ 𝑥 + 𝑏
What is a Machine Learning Model?
Machine Learning with Neural Networks, TensorFlow and Keras13 9/26/2018
Machine Learning Model
predict
Architecture
(Linear, Polynomial, Decision Tree, SVM, NN, …)
Parameters Hyper Parameters
(model config, training)
Training
Training data:
features + labels
Cost Function
Machine Learning with Neural Networks, TensorFlow and Keras14 9/26/2018
How good is our model in predicting labels for a given data set?
Calculate the distance from the predicted to the actual labels for all values of x and
sum them up “somehow”
– For regression often the “mean squared error” (MSE) is used
Model
Parameters
Minimize Cost Function with “Gradient Decent”
Machine Learning with Neural Networks, TensorFlow and Keras15 9/26/2018
x
y
Constant Model
Parameters
(b = 2000000)
𝑦 = 𝑏
b
J(b)
𝑏 𝑛+1 = 𝑏 𝑛 − 𝛼
𝜕
𝜕𝑏 𝑛
𝐽(𝑏 𝑛)
Learning
Rate
𝑏
Gradient Decent & Friends with 2 Parameters
Machine Learning with Neural Networks, TensorFlow and Keras16 9/26/2018
http://www.denizyuret.com/2015/03/alec-radfords-animations-for.html
Under- and Over-Fitting, Validation Set
Machine Learning with Neural Networks, TensorFlow and Keras17 9/26/2018
x
y
Linear
Constant
Super-duper
Put some data aside for validation
Training
Set (~80%)
Validation
Set (~20%)
Machine Learning with Neural Networks, TensorFlow and Keras18 9/26/2018
Neural Networks
Machine Learning with Neural Networks, TensorFlow and Keras19 9/26/2018
𝑥1
𝑥2
𝑥3
𝑥4
∗ 𝑤1
1
∗ 𝑤2
1
∗ 𝑤3
1
∗ 𝑤4
1
𝑏1
𝑧1
∗ 𝑤1
2
∗ 𝑤2
2
∗ 𝑤3
2
∗ 𝑤4
2
𝑏2
𝑧2
∗ 𝑤1
3
∗ 𝑤2
3
∗ 𝑤3
3
∗ 𝑤4
3
𝑏3
𝑧3
Softmax(ActivationFunction)
𝑦1
𝑦2
𝑦3
Neuron 1
Neuron 2
Neuron 3 𝑦𝑖 = 1
Use with
“categorical crossentropy”
cost function
Neural Network Layers
Machine Learning with Neural Networks, TensorFlow and Keras20 9/26/2018
input
layer
output
layer
Softmax
Simpler representation:
Hidden Layers
Machine Learning with Neural Networks, TensorFlow and Keras21 9/26/2018
SoftmaxReLU
ReLU
Activation Functions
Machine Learning with Neural Networks, TensorFlow and Keras22 9/26/2018
https://medium.com/@shrutijadon10104776/survey-on-activation-functions-for-deep-learning-9689331ba092
Training
Machine Learning with Neural Networks, TensorFlow and Keras23 9/26/2018
activation
activation
activation
activation
X YY
Cost func
Forward pass
Backward pass
Backward pass:
• calculate gradients for all parameters (weights, biases)
• Perform one step of gradient decent
Machine Learning with Neural Networks, TensorFlow and Keras24 9/26/2018
Using TensorFlow and Keras
TensorFlow / Keras
Machine Learning with Neural Networks, TensorFlow and Keras25 9/26/2018
What is a Tensor?
– Generalization of Scalars  Rank 0, Vectors  Rank 1, Matrixes  Rank 2, …
TensorFlow
– Framework to perform mathematical calculations with tensors
– You build graphs of expression and execute them in a TensorFlow session
– Written in C++, language bindings to Python and other languages
Keras
– High level API over TensorFlow, CNTK and Theano
– More declarative
– Written in Python
Building a neural network with Keras
Machine Learning with Neural Networks, TensorFlow and Keras26 9/26/2018
def build_model():
from tensorflow.keras.layers import Dense, Input
from tensorflow.keras.models import Model
input = Input(shape=(4,), name="input_layer")
x = Dense(10, activation='relu', name="hidden_layer_1")(input)
x = Dense(8, activation='relu', name="hidden_layer_2")(x)
x = Dense(6, activation='relu', name="hidden_layer_3")(x)
output = Dense(3, activation='softmax', name="output_layer")(x)
model = Model(inputs=input, outputs=output)
return model
Setting up Training
Machine Learning with Neural Networks, TensorFlow and Keras27 9/26/2018
def compile_model(model):
from tensorflow.keras.optimizers import Adam
optimizer = Adam(lr=0.04, decay=1e-7)
lossfunction = 'categorical_crossentropy'
model.compile(optimizer,
lossfunction,
metrics=['accuracy'])
Training the model
Machine Learning with Neural Networks, TensorFlow and Keras28 9/26/2018
model.fit(training_dataset,
steps_per_epoch=TRAINING_STEPS,
epochs=50)
model.save(".modeliris.h5")
Making Predictions
Machine Learning with Neural Networks, TensorFlow and Keras29 9/26/2018
import numpy as np
from tensorflow.keras.models import load_model
code_to_name = {0: "setosa", 1: "versicolor", 2: "virginica"}
model = load_model(".modeliris.h5")
def predict(features):
x = np.array([features])
y = model.predict([x])
print("Model output: {}".format(y))
code = np.argmax(y)
return code_to_name[code]
prediction = predict([5.9, 3, 4.2, 1.5])
print("Prediction: {}".format(prediction))
Useful Tools
Machine Learning with Neural Networks, TensorFlow and Keras30 9/26/2018
Lutz Roeder’s Netron
– Visualizes saved models
– https://github.com/lutzroeder/Netron
TensorBoard
– Visualizes output from the
training process
– Part of TensorFlow
Using the Model in a .NET Application
Machine Learning with Neural Networks, TensorFlow and Keras31 9/26/2018
Save the model as a TensorFlow checkpoint
Freeze the saved model, save it as a .pb file
– Using the script: tensorflowpythontoolsfreeze_graph.py
Use TensorFlowSharp to load the model and make predictions
– https://github.com/migueldeicaza/TensorFlowSharp
from tensorflow.keras import backend as K
saver = tf.train.Saver()
saver.save(K.get_session(), '.modelkeras_model.ckpt')
Using the Model in a .NET Application
Machine Learning with Neural Networks, TensorFlow and Keras32 9/26/2018
public static IrisPrediction Predict(IrisFeatures features)
{
var data = features.ToArray();
var tensor = TFTensor.FromBuffer(new TFShape(1, data.Length), data, 0, data.Length);
using (var graph = new TFGraph())
{
graph.Import(File.ReadAllBytes("keras_frozen.pb"));
var session = new TFSession(graph);
var runner = session.GetRunner();
runner.AddInput(graph["input_layer"][0], tensor);
runner.Fetch(graph["output_layer/Softmax"][0]);
var output = runner.Run();
TFTensor result = output[0];
float[] p = ((float[][])result.GetValue(true))[0];
return IrisPrediction.FromArray(p);
}
}
Last Words
Machine Learning with Neural Networks, TensorFlow and Keras33 9/26/2018
Know the ML terminology
There are many ways to do things
– Start with one
This was only an introduction
– There are much more possibilities
Reuse what has been done by others
– Network designs
– Transfer learning
Name Presentation34 9/26/2018
Sample Code: https://github.com/cpletz/TFKerasTEDemo
Session Feedback – now
TechEvent September 201835 14.09.2018
Please use the Trivadis Events mobile app to give feedback on each session
Use "My schedule" if you have registered for a session
Otherwise use "Agenda" and the search function
If the mobile app does not work (or if you have a Windows smartphone), use your
smartphone browser
– URL: http://trivadis.quickmobileplatform.eu/
– User name: <your_loginname> (such as "svv")
– Password: sent by e-mail...
Christoph Pletz
Principal Consultant
Tel. +41 79 457 97 31
christoph.pletz@trivadis.com
9/26/2018 Name Presentation36

Mais conteúdo relacionado

Mais procurados

An introduction to deep learning
An introduction to deep learningAn introduction to deep learning
An introduction to deep learningVan Thanh
 
Matching networks for one shot learning
Matching networks for one shot learningMatching networks for one shot learning
Matching networks for one shot learningKazuki Fujikawa
 
Google Developer Groups Talk - TensorFlow
Google Developer Groups Talk - TensorFlowGoogle Developer Groups Talk - TensorFlow
Google Developer Groups Talk - TensorFlowHarini Gunabalan
 
Machine learning in science and industry — day 2
Machine learning in science and industry — day 2Machine learning in science and industry — day 2
Machine learning in science and industry — day 2arogozhnikov
 
Dictionary Learning for Massive Matrix Factorization
Dictionary Learning for Massive Matrix FactorizationDictionary Learning for Massive Matrix Factorization
Dictionary Learning for Massive Matrix Factorizationrecsysfr
 
Machine learning and_neural_network_lecture_slide_ece_dku
Machine learning and_neural_network_lecture_slide_ece_dkuMachine learning and_neural_network_lecture_slide_ece_dku
Machine learning and_neural_network_lecture_slide_ece_dkuSeokhyun Yoon
 
Summarizing videos with Attention
Summarizing videos with AttentionSummarizing videos with Attention
Summarizing videos with AttentionArithmer Inc.
 
10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systemsXavier Amatriain
 
Using Deep Learning to Find Similar Dresses
Using Deep Learning to Find Similar DressesUsing Deep Learning to Find Similar Dresses
Using Deep Learning to Find Similar DressesHJ van Veen
 
Prototype-based models in machine learning
Prototype-based models in machine learningPrototype-based models in machine learning
Prototype-based models in machine learningUniversity of Groningen
 
Neural netorksmatching
Neural netorksmatchingNeural netorksmatching
Neural netorksmatchingMasa Kato
 
Mahoney mlconf-nov13
Mahoney mlconf-nov13Mahoney mlconf-nov13
Mahoney mlconf-nov13MLconf
 
Machine learning in science and industry — day 1
Machine learning in science and industry — day 1Machine learning in science and industry — day 1
Machine learning in science and industry — day 1arogozhnikov
 
Neural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronNeural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronMostafa G. M. Mostafa
 
Introduction To TensorFlow
Introduction To TensorFlowIntroduction To TensorFlow
Introduction To TensorFlowSpotle.ai
 
Introduction to Machine Learning with Python and scikit-learn
Introduction to Machine Learning with Python and scikit-learnIntroduction to Machine Learning with Python and scikit-learn
Introduction to Machine Learning with Python and scikit-learnMatt Hagy
 
Introduction to Tensor Flow for Optical Character Recognition (OCR)
Introduction to Tensor Flow for Optical Character Recognition (OCR)Introduction to Tensor Flow for Optical Character Recognition (OCR)
Introduction to Tensor Flow for Optical Character Recognition (OCR)Vincenzo Santopietro
 
SVM Tutorial
SVM TutorialSVM Tutorial
SVM Tutorialbutest
 

Mais procurados (20)

Bol.com
Bol.comBol.com
Bol.com
 
An introduction to deep learning
An introduction to deep learningAn introduction to deep learning
An introduction to deep learning
 
Tensor flow
Tensor flowTensor flow
Tensor flow
 
Matching networks for one shot learning
Matching networks for one shot learningMatching networks for one shot learning
Matching networks for one shot learning
 
Google Developer Groups Talk - TensorFlow
Google Developer Groups Talk - TensorFlowGoogle Developer Groups Talk - TensorFlow
Google Developer Groups Talk - TensorFlow
 
Machine learning in science and industry — day 2
Machine learning in science and industry — day 2Machine learning in science and industry — day 2
Machine learning in science and industry — day 2
 
Dictionary Learning for Massive Matrix Factorization
Dictionary Learning for Massive Matrix FactorizationDictionary Learning for Massive Matrix Factorization
Dictionary Learning for Massive Matrix Factorization
 
Machine learning and_neural_network_lecture_slide_ece_dku
Machine learning and_neural_network_lecture_slide_ece_dkuMachine learning and_neural_network_lecture_slide_ece_dku
Machine learning and_neural_network_lecture_slide_ece_dku
 
Summarizing videos with Attention
Summarizing videos with AttentionSummarizing videos with Attention
Summarizing videos with Attention
 
10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems
 
Using Deep Learning to Find Similar Dresses
Using Deep Learning to Find Similar DressesUsing Deep Learning to Find Similar Dresses
Using Deep Learning to Find Similar Dresses
 
Prototype-based models in machine learning
Prototype-based models in machine learningPrototype-based models in machine learning
Prototype-based models in machine learning
 
Neural netorksmatching
Neural netorksmatchingNeural netorksmatching
Neural netorksmatching
 
Mahoney mlconf-nov13
Mahoney mlconf-nov13Mahoney mlconf-nov13
Mahoney mlconf-nov13
 
Machine learning in science and industry — day 1
Machine learning in science and industry — day 1Machine learning in science and industry — day 1
Machine learning in science and industry — day 1
 
Neural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronNeural Networks: Multilayer Perceptron
Neural Networks: Multilayer Perceptron
 
Introduction To TensorFlow
Introduction To TensorFlowIntroduction To TensorFlow
Introduction To TensorFlow
 
Introduction to Machine Learning with Python and scikit-learn
Introduction to Machine Learning with Python and scikit-learnIntroduction to Machine Learning with Python and scikit-learn
Introduction to Machine Learning with Python and scikit-learn
 
Introduction to Tensor Flow for Optical Character Recognition (OCR)
Introduction to Tensor Flow for Optical Character Recognition (OCR)Introduction to Tensor Flow for Optical Character Recognition (OCR)
Introduction to Tensor Flow for Optical Character Recognition (OCR)
 
SVM Tutorial
SVM TutorialSVM Tutorial
SVM Tutorial
 

Semelhante a TechEvent Machine Learning

IRJET- Automatic Object Sorting using Deep Learning
IRJET- Automatic Object Sorting using Deep LearningIRJET- Automatic Object Sorting using Deep Learning
IRJET- Automatic Object Sorting using Deep LearningIRJET Journal
 
Google Big Data Expo
Google Big Data ExpoGoogle Big Data Expo
Google Big Data ExpoBigDataExpo
 
House price prediction
House price predictionHouse price prediction
House price predictionSabahBegum
 
DLT UNIT-3.docx
DLT  UNIT-3.docxDLT  UNIT-3.docx
DLT UNIT-3.docx0567Padma
 
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARN
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARNMLConf 2013: Metronome and Parallel Iterative Algorithms on YARN
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARNJosh Patterson
 
maxbox starter60 machine learning
maxbox starter60 machine learningmaxbox starter60 machine learning
maxbox starter60 machine learningMax Kleiner
 
A TALE of DATA PATTERN DISCOVERY IN PARALLEL
A TALE of DATA PATTERN DISCOVERY IN PARALLELA TALE of DATA PATTERN DISCOVERY IN PARALLEL
A TALE of DATA PATTERN DISCOVERY IN PARALLELJenny Liu
 
Josh Patterson MLconf slides
Josh Patterson MLconf slidesJosh Patterson MLconf slides
Josh Patterson MLconf slidesMLconf
 
Web Traffic Time Series Forecasting
Web Traffic  Time Series ForecastingWeb Traffic  Time Series Forecasting
Web Traffic Time Series ForecastingBillTubbs
 
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud ML
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud MLScaling TensorFlow Models for Training using multi-GPUs & Google Cloud ML
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud MLSeldon
 
Nexxworks bootcamp ML6 (27/09/2017)
Nexxworks bootcamp ML6 (27/09/2017)Nexxworks bootcamp ML6 (27/09/2017)
Nexxworks bootcamp ML6 (27/09/2017)Karel Dumon
 
lecture-intro-pet-nams-ai-in-toxicology.pptx
lecture-intro-pet-nams-ai-in-toxicology.pptxlecture-intro-pet-nams-ai-in-toxicology.pptx
lecture-intro-pet-nams-ai-in-toxicology.pptxMarc Teunis
 
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnIntroduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnBenjamin Bengfort
 
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...inside-BigData.com
 
Intro to Deep Learning with Keras - using TensorFlow backend
Intro to Deep Learning with Keras - using TensorFlow backendIntro to Deep Learning with Keras - using TensorFlow backend
Intro to Deep Learning with Keras - using TensorFlow backendAmin Golnari
 
Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for BeginnersSanghamitra Deb
 
Presentation on BornoNet Research Paper and Python Basics
Presentation on BornoNet Research Paper and Python BasicsPresentation on BornoNet Research Paper and Python Basics
Presentation on BornoNet Research Paper and Python BasicsShibbir Ahmed
 
Dog Breed Classification using PyTorch on Azure Machine Learning
Dog Breed Classification using PyTorch on Azure Machine LearningDog Breed Classification using PyTorch on Azure Machine Learning
Dog Breed Classification using PyTorch on Azure Machine LearningHeather Spetalnick
 

Semelhante a TechEvent Machine Learning (20)

IRJET- Automatic Object Sorting using Deep Learning
IRJET- Automatic Object Sorting using Deep LearningIRJET- Automatic Object Sorting using Deep Learning
IRJET- Automatic Object Sorting using Deep Learning
 
Google Big Data Expo
Google Big Data ExpoGoogle Big Data Expo
Google Big Data Expo
 
House price prediction
House price predictionHouse price prediction
House price prediction
 
C3 w1
C3 w1C3 w1
C3 w1
 
DLT UNIT-3.docx
DLT  UNIT-3.docxDLT  UNIT-3.docx
DLT UNIT-3.docx
 
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARN
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARNMLConf 2013: Metronome and Parallel Iterative Algorithms on YARN
MLConf 2013: Metronome and Parallel Iterative Algorithms on YARN
 
maxbox starter60 machine learning
maxbox starter60 machine learningmaxbox starter60 machine learning
maxbox starter60 machine learning
 
A TALE of DATA PATTERN DISCOVERY IN PARALLEL
A TALE of DATA PATTERN DISCOVERY IN PARALLELA TALE of DATA PATTERN DISCOVERY IN PARALLEL
A TALE of DATA PATTERN DISCOVERY IN PARALLEL
 
Josh Patterson MLconf slides
Josh Patterson MLconf slidesJosh Patterson MLconf slides
Josh Patterson MLconf slides
 
Web Traffic Time Series Forecasting
Web Traffic  Time Series ForecastingWeb Traffic  Time Series Forecasting
Web Traffic Time Series Forecasting
 
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud ML
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud MLScaling TensorFlow Models for Training using multi-GPUs & Google Cloud ML
Scaling TensorFlow Models for Training using multi-GPUs & Google Cloud ML
 
Nexxworks bootcamp ML6 (27/09/2017)
Nexxworks bootcamp ML6 (27/09/2017)Nexxworks bootcamp ML6 (27/09/2017)
Nexxworks bootcamp ML6 (27/09/2017)
 
lecture-intro-pet-nams-ai-in-toxicology.pptx
lecture-intro-pet-nams-ai-in-toxicology.pptxlecture-intro-pet-nams-ai-in-toxicology.pptx
lecture-intro-pet-nams-ai-in-toxicology.pptx
 
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnIntroduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-Learn
 
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
 
Intro to Deep Learning with Keras - using TensorFlow backend
Intro to Deep Learning with Keras - using TensorFlow backendIntro to Deep Learning with Keras - using TensorFlow backend
Intro to Deep Learning with Keras - using TensorFlow backend
 
Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for Beginners
 
Presentation on BornoNet Research Paper and Python Basics
Presentation on BornoNet Research Paper and Python BasicsPresentation on BornoNet Research Paper and Python Basics
Presentation on BornoNet Research Paper and Python Basics
 
Dog Breed Classification using PyTorch on Azure Machine Learning
Dog Breed Classification using PyTorch on Azure Machine LearningDog Breed Classification using PyTorch on Azure Machine Learning
Dog Breed Classification using PyTorch on Azure Machine Learning
 
Deep learning-practical
Deep learning-practicalDeep learning-practical
Deep learning-practical
 

Mais de Trivadis

Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...
Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...
Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...Trivadis
 
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...Trivadis
 
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)Trivadis
 
Azure Days 2019: Master the Move to Azure (Konrad Brunner)
Azure Days 2019: Master the Move to Azure (Konrad Brunner)Azure Days 2019: Master the Move to Azure (Konrad Brunner)
Azure Days 2019: Master the Move to Azure (Konrad Brunner)Trivadis
 
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...Trivadis
 
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)Trivadis
 
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...Trivadis
 
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...Trivadis
 
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...Trivadis
 
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...Trivadis
 
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...Trivadis
 
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...Trivadis
 
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - Trivadis
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - TrivadisTechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - Trivadis
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - TrivadisTrivadis
 
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...Trivadis
 
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...Trivadis
 
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...Trivadis
 
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...Trivadis
 
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...Trivadis
 
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...Trivadis
 
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - Trivadis
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - TrivadisTechEvent 2019: The sleeping Power of Data; Eberhard Lösch - Trivadis
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - TrivadisTrivadis
 

Mais de Trivadis (20)

Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...
Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...
Azure Days 2019: Azure Chatbot Development for Airline Irregularities (Remco ...
 
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...
Azure Days 2019: Trivadis Azure Foundation – Das Fundament für den ... (Nisan...
 
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)
Azure Days 2019: Business Intelligence auf Azure (Marco Amhof & Yves Mauron)
 
Azure Days 2019: Master the Move to Azure (Konrad Brunner)
Azure Days 2019: Master the Move to Azure (Konrad Brunner)Azure Days 2019: Master the Move to Azure (Konrad Brunner)
Azure Days 2019: Master the Move to Azure (Konrad Brunner)
 
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...
Azure Days 2019: Keynote Azure Switzerland – Status Quo und Ausblick (Primo A...
 
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)
Azure Days 2019: Grösser und Komplexer ist nicht immer besser (Meinrad Weiss)
 
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...
Azure Days 2019: Get Connected with Azure API Management (Gerry Keune & Stefa...
 
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...
Azure Days 2019: Infrastructure as Code auf Azure (Jonas Wanninger & Daniel H...
 
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...
Azure Days 2019: Wie bringt man eine Data Analytics Plattform in die Cloud? (...
 
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...
Azure Days 2019: Azure@Helsana: Die Erweiterung von Dynamics CRM mit Azure Po...
 
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...
TechEvent 2019: Kundenstory - Kein Angebot, kein Auftrag – Wie Du ein individ...
 
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...
TechEvent 2019: Oracle Database Appliance M/L - Erfahrungen und Erfolgsmethod...
 
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - Trivadis
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - TrivadisTechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - Trivadis
TechEvent 2019: Security 101 für Web Entwickler; Roland Krüger - Trivadis
 
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...
TechEvent 2019: Trivadis & Swisscom Partner Angebote; Konrad Häfeli, Markus O...
 
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...
TechEvent 2019: DBaaS from Swisscom Cloud powered by Trivadis; Konrad Häfeli ...
 
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...
TechEvent 2019: Status of the partnership Trivadis and EDB - Comparing Postgr...
 
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...
TechEvent 2019: More Agile, More AI, More Cloud! Less Work?!; Oliver Dörr - T...
 
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...
TechEvent 2019: Kundenstory - Vom Hauptmann zu Köpenick zum Polizisten 2020 -...
 
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...
TechEvent 2019: Vom Rechenzentrum in die Oracle Cloud - Übertragungsmethoden;...
 
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - Trivadis
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - TrivadisTechEvent 2019: The sleeping Power of Data; Eberhard Lösch - Trivadis
TechEvent 2019: The sleeping Power of Data; Eberhard Lösch - Trivadis
 

Último

#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Allon Mureinik
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 

Último (20)

#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 

TechEvent Machine Learning

  • 1. BASEL BERN BRUGG DÜSSELDORF FRANKFURT A.M. FREIBURG I.BR. GENEVA HAMBURG COPENHAGEN LAUSANNE MUNICH STUTTGART VIENNA ZURICH Machine Learning … … with Neural Networks, TensorFlow and Keras Christoph Pletz
  • 2. Agenda Name Presentation2 9/26/2018 1. Machine Learning Basics 2. Neural Networks 3. Using TensorFlow and Keras
  • 3. Machine Learning with Neural Networks, TensorFlow and Keras3 9/26/2018 Machine Learning Basics
  • 4. Types of Machine Learning Machine Learning with Neural Networks, TensorFlow and Keras4 9/26/2018 Regression – How much? How many? Classification – Which class does this belong to? Clustering – Are there different groups? To which group does an item belong? Anomaly Detection – Is this weird? Recommendation – Which option to choose? supervisedunsupervised
  • 5. Simplest Example – housing prices Machine Learning with Neural Networks, TensorFlow and Keras5 9/26/2018 Living Space Price Living Space Price 95 1000000 100 950000 120 1200000 … … 250 3300000 Given a house with a certain living space, what is the expected price?
  • 6. Terminology: Features and Labels Machine Learning with Neural Networks, TensorFlow and Keras6 9/26/2018 Living Space Price 95 1000000 100 950000 120 1200000 … … 250 3300000 Feature(s) x Label(s) y
  • 7. Hypothesis Function / Model Machine Learning with Neural Networks, TensorFlow and Keras7 9/26/2018 𝑦 = ℎ(𝑥) hypothesis model 𝑥 predict 𝑦 𝑦 𝑦 Actual label value Predicted value 𝑥 𝑦 𝑦 95 1000000 998734 100 950000 1043576 120 1200000 1236556 … …
  • 8. Second Simplest Example – Iris classification Machine Learning with Neural Networks, TensorFlow and Keras8 9/26/2018 CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=170298 Iris setosa Iris versicolor Iris virginica Data collected by Edgar Anderson and published by Ronald Fisher in 1936 (150 samples)
  • 9. Iris Data Set Machine Learning with Neural Networks, TensorFlow and Keras9 9/26/2018 https://upload.wikimedia.org/wikipedia/commons/thumb/5/56/Iris_dataset_scatterplot.svg/749px-Iris_dataset_scatterplot.svg.png
  • 10. Iris Data Set (classification) Machine Learning with Neural Networks, TensorFlow and Keras10 9/26/2018 Sepal length Sepal width Petal length Petal width Species 5.9 3.0 4.2 1.5 versicolor 6.9 3.1 5.4 2.1 virginica 5.1 3.3 1.7 0.5 setosa … … … … … Species code 1 2 0 … Is Setosa Is versicolor Is virginica 0 1 0 0 0 1 1 0 0 … … … One hot encoding Features x Labels y
  • 11. A bit more challenging: Image Recognition Machine Learning with Neural Networks, TensorFlow and Keras11 9/26/2018 Lets assume we want to classify the iris species from images … E.g. 512 * 512 pixels and 3 color channels  786432 features, still 3 classes Imagenet (http://www.image-net.org/) provides pictures for 1000 classes model predict 𝑦
  • 12. A Linear Model for Housing Prices Machine Learning with Neural Networks, TensorFlow and Keras12 9/26/2018 x y Linear Model Parameters (b = 100000, w = 0.5) 𝑦 = 𝑤 ∗ 𝑥 + 𝑏
  • 13. What is a Machine Learning Model? Machine Learning with Neural Networks, TensorFlow and Keras13 9/26/2018 Machine Learning Model predict Architecture (Linear, Polynomial, Decision Tree, SVM, NN, …) Parameters Hyper Parameters (model config, training) Training Training data: features + labels
  • 14. Cost Function Machine Learning with Neural Networks, TensorFlow and Keras14 9/26/2018 How good is our model in predicting labels for a given data set? Calculate the distance from the predicted to the actual labels for all values of x and sum them up “somehow” – For regression often the “mean squared error” (MSE) is used Model Parameters
  • 15. Minimize Cost Function with “Gradient Decent” Machine Learning with Neural Networks, TensorFlow and Keras15 9/26/2018 x y Constant Model Parameters (b = 2000000) 𝑦 = 𝑏 b J(b) 𝑏 𝑛+1 = 𝑏 𝑛 − 𝛼 𝜕 𝜕𝑏 𝑛 𝐽(𝑏 𝑛) Learning Rate 𝑏
  • 16. Gradient Decent & Friends with 2 Parameters Machine Learning with Neural Networks, TensorFlow and Keras16 9/26/2018 http://www.denizyuret.com/2015/03/alec-radfords-animations-for.html
  • 17. Under- and Over-Fitting, Validation Set Machine Learning with Neural Networks, TensorFlow and Keras17 9/26/2018 x y Linear Constant Super-duper Put some data aside for validation Training Set (~80%) Validation Set (~20%)
  • 18. Machine Learning with Neural Networks, TensorFlow and Keras18 9/26/2018 Neural Networks
  • 19. Machine Learning with Neural Networks, TensorFlow and Keras19 9/26/2018 𝑥1 𝑥2 𝑥3 𝑥4 ∗ 𝑤1 1 ∗ 𝑤2 1 ∗ 𝑤3 1 ∗ 𝑤4 1 𝑏1 𝑧1 ∗ 𝑤1 2 ∗ 𝑤2 2 ∗ 𝑤3 2 ∗ 𝑤4 2 𝑏2 𝑧2 ∗ 𝑤1 3 ∗ 𝑤2 3 ∗ 𝑤3 3 ∗ 𝑤4 3 𝑏3 𝑧3 Softmax(ActivationFunction) 𝑦1 𝑦2 𝑦3 Neuron 1 Neuron 2 Neuron 3 𝑦𝑖 = 1 Use with “categorical crossentropy” cost function
  • 20. Neural Network Layers Machine Learning with Neural Networks, TensorFlow and Keras20 9/26/2018 input layer output layer Softmax Simpler representation:
  • 21. Hidden Layers Machine Learning with Neural Networks, TensorFlow and Keras21 9/26/2018 SoftmaxReLU ReLU
  • 22. Activation Functions Machine Learning with Neural Networks, TensorFlow and Keras22 9/26/2018 https://medium.com/@shrutijadon10104776/survey-on-activation-functions-for-deep-learning-9689331ba092
  • 23. Training Machine Learning with Neural Networks, TensorFlow and Keras23 9/26/2018 activation activation activation activation X YY Cost func Forward pass Backward pass Backward pass: • calculate gradients for all parameters (weights, biases) • Perform one step of gradient decent
  • 24. Machine Learning with Neural Networks, TensorFlow and Keras24 9/26/2018 Using TensorFlow and Keras
  • 25. TensorFlow / Keras Machine Learning with Neural Networks, TensorFlow and Keras25 9/26/2018 What is a Tensor? – Generalization of Scalars  Rank 0, Vectors  Rank 1, Matrixes  Rank 2, … TensorFlow – Framework to perform mathematical calculations with tensors – You build graphs of expression and execute them in a TensorFlow session – Written in C++, language bindings to Python and other languages Keras – High level API over TensorFlow, CNTK and Theano – More declarative – Written in Python
  • 26. Building a neural network with Keras Machine Learning with Neural Networks, TensorFlow and Keras26 9/26/2018 def build_model(): from tensorflow.keras.layers import Dense, Input from tensorflow.keras.models import Model input = Input(shape=(4,), name="input_layer") x = Dense(10, activation='relu', name="hidden_layer_1")(input) x = Dense(8, activation='relu', name="hidden_layer_2")(x) x = Dense(6, activation='relu', name="hidden_layer_3")(x) output = Dense(3, activation='softmax', name="output_layer")(x) model = Model(inputs=input, outputs=output) return model
  • 27. Setting up Training Machine Learning with Neural Networks, TensorFlow and Keras27 9/26/2018 def compile_model(model): from tensorflow.keras.optimizers import Adam optimizer = Adam(lr=0.04, decay=1e-7) lossfunction = 'categorical_crossentropy' model.compile(optimizer, lossfunction, metrics=['accuracy'])
  • 28. Training the model Machine Learning with Neural Networks, TensorFlow and Keras28 9/26/2018 model.fit(training_dataset, steps_per_epoch=TRAINING_STEPS, epochs=50) model.save(".modeliris.h5")
  • 29. Making Predictions Machine Learning with Neural Networks, TensorFlow and Keras29 9/26/2018 import numpy as np from tensorflow.keras.models import load_model code_to_name = {0: "setosa", 1: "versicolor", 2: "virginica"} model = load_model(".modeliris.h5") def predict(features): x = np.array([features]) y = model.predict([x]) print("Model output: {}".format(y)) code = np.argmax(y) return code_to_name[code] prediction = predict([5.9, 3, 4.2, 1.5]) print("Prediction: {}".format(prediction))
  • 30. Useful Tools Machine Learning with Neural Networks, TensorFlow and Keras30 9/26/2018 Lutz Roeder’s Netron – Visualizes saved models – https://github.com/lutzroeder/Netron TensorBoard – Visualizes output from the training process – Part of TensorFlow
  • 31. Using the Model in a .NET Application Machine Learning with Neural Networks, TensorFlow and Keras31 9/26/2018 Save the model as a TensorFlow checkpoint Freeze the saved model, save it as a .pb file – Using the script: tensorflowpythontoolsfreeze_graph.py Use TensorFlowSharp to load the model and make predictions – https://github.com/migueldeicaza/TensorFlowSharp from tensorflow.keras import backend as K saver = tf.train.Saver() saver.save(K.get_session(), '.modelkeras_model.ckpt')
  • 32. Using the Model in a .NET Application Machine Learning with Neural Networks, TensorFlow and Keras32 9/26/2018 public static IrisPrediction Predict(IrisFeatures features) { var data = features.ToArray(); var tensor = TFTensor.FromBuffer(new TFShape(1, data.Length), data, 0, data.Length); using (var graph = new TFGraph()) { graph.Import(File.ReadAllBytes("keras_frozen.pb")); var session = new TFSession(graph); var runner = session.GetRunner(); runner.AddInput(graph["input_layer"][0], tensor); runner.Fetch(graph["output_layer/Softmax"][0]); var output = runner.Run(); TFTensor result = output[0]; float[] p = ((float[][])result.GetValue(true))[0]; return IrisPrediction.FromArray(p); } }
  • 33. Last Words Machine Learning with Neural Networks, TensorFlow and Keras33 9/26/2018 Know the ML terminology There are many ways to do things – Start with one This was only an introduction – There are much more possibilities Reuse what has been done by others – Network designs – Transfer learning
  • 34. Name Presentation34 9/26/2018 Sample Code: https://github.com/cpletz/TFKerasTEDemo
  • 35. Session Feedback – now TechEvent September 201835 14.09.2018 Please use the Trivadis Events mobile app to give feedback on each session Use "My schedule" if you have registered for a session Otherwise use "Agenda" and the search function If the mobile app does not work (or if you have a Windows smartphone), use your smartphone browser – URL: http://trivadis.quickmobileplatform.eu/ – User name: <your_loginname> (such as "svv") – Password: sent by e-mail...
  • 36. Christoph Pletz Principal Consultant Tel. +41 79 457 97 31 christoph.pletz@trivadis.com 9/26/2018 Name Presentation36