Add support for Mobilenet v2 in Image Classification transfer learning #4351
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
eerhardt
reviewed
Oct 18, 2019
| @@ -1249,7 +1249,7 @@ public void TensorFlowImageClassification() | |||
| var pipeline = mlContext.Transforms.LoadImages("Image", fullImagesetFolderPath, false, "ImagePath") // false indicates we want the image as a VBuffer<byte> | |||
| .Append(mlContext.Model.ImageClassification( | |||
| "Image", "Label", | |||
| // Just by changing/selecting InceptionV3 here instead of | |||
| // Just by changing/selecting InceptionV3/MobilenetV2 here instead of | |||
There was a problem hiding this comment.
Do you want to add a test for the new functionality? The CI isn't executing any of your new code. #Resolved
There was a problem hiding this comment.
Codecov Report
@@ Coverage Diff @@
## master #4351 +/- ##
==========================================
+ Coverage 74.59% 74.6% +<.01%
==========================================
Files 881 882 +1
Lines 154815 154848 +33
Branches 16910 16913 +3
==========================================
+ Hits 115486 115524 +38
+ Misses 34578 34574 -4
+ Partials 4751 4750 -1
|
codemzs
reviewed
Oct 21, 2019
frank-dong-ms-zz
pushed a commit
to frank-dong-ms-zz/machinelearning
that referenced
this pull request
Nov 4, 2019
dotnet#4351) * add Mobilenet_V2 architecture support and sample * Separated try catch fix in different issue, added model download location * removed redundant sample, changed comment message to add MobilenetV2 * added test for MobileNetV2, reusing test with Theory
ghost
locked as resolved and limited conversation to collaborators
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The base MobileNetV2 model is from https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet . Checkpoint file used: https://storage.googleapis.com/mobilenet_v2/checkpoints/mobilenet_v2_1.4_224.tgz
Benchmarked transfer learning on the Cats V/s Dogs dataset at:
https://www.microsoft.com/en-us/download/details.aspx?id=54765
Trained last layer for 10 epochs with 0.0001f learning rate with batch size 32.
Final accuracy on eval set:
Micro-accuracy: 0.980959936533122,macro-accuracy = 0.980874311779911
Comparable Tensorflow acuracy for same task:
https://www.tensorflow.org/tutorials/images/transfer_learning#train_the_model