What's Inside the Box?

A presentation at PAPIs.io in October 2018 in Cambridge, MA, USA by Ray Deck

Slide 1

What’s Inside the Box? Comparing Data Storage Across Frameworks @ray_deck

Slide 2

Slide 3

Code as Data

Slide 4

PyTorch Keras TensorFlow TensorFlow Lite CoreML

Slide 5

Slide 6

Serialization pytorch.org/ docs/ stable/ notes/ serialization.html

Slide 7

Pickle docs.python.org/3/library/pickle.html

Slide 8

Slide 9

JSON + HDF5

Slide 10

Slide 11

Slide 12

HDFView portal.hdfgroup.org/display/support/Download+HDFView Panoply www.giss.nasa.gov/tools/panoply/

Slide 13

JSON in HDF5

Slide 14

https://sparktoro.com/fake-followers/ray_deck

Slide 15

Slide 16

Slide 17

developers.google.com/protocol-buffers

Slide 18

https://www.slideshare.net/SergeyPodolsky/google-protocol-buffers-56085699

Slide 19

github.com/ tensorflow/ tensorflow/ tree/ master/ tensorflow/ core/ protobuf

Slide 20

Slide 21

Protobuf Viewer MacOS App Store Protobuf Editor sourceforge.net/projects/protobufeditor/

Slide 22

Slide 23

Slide 24

FlatBuf google.github.io/ flatbuffers/

Slide 25

FlatBuf github.com/ tensorflow/ tensorflow/ blob/ master/ tensorflow/ contrib/ lite/ schema/ schema_v3.fbs

Slide 26

Slide 27

github.com/ apple/ coremltools/ tree/ master/ mlmodel/ format

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

let c = MLModel.compile(url: u) let m = MLModel(url: c) let d:[String:Any] = ["input": image] let dfpin = MLDictionaryFeatureProvider(dictionary: d) let fp = m.predict(dfp) let dfpout = fp.featureValue(for: "labels") let dic = dfpout.dictionaryValue let topResult = dic.first()

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Quo Vadis?

Slide 40

Slide 41

Thank You @ray_deck github.com/rhdeck/papis-2018