-
Notifications
You must be signed in to change notification settings - Fork 0
vivek7266/wootz
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Wootz Compiler Code Structure: Grammar: src/main/java/grammars/CaffePrototxt.g4 ANTLR generated files: src/main/java/gen Test cases: src/main/resources/*.prototxt output: *.py Main class: src/main/java/wootz/WootzMain.java Layer Generators: src/main/java/generators Software requirements: ANTLR v4 Apache Maven 3.5.4 Java version >= 8 (this project also runs on java 10.0.2 ) Running instructions: In the home directory: wootz this should be the working directory for following command to work. $ mvn clean install $ java -jar target/wootz-1.0-SNAPSHOT.jar <path_of_prototxt> <multiplexing_boolean> For example: =================================================================================================== 1. To run inception_v1.prototxt without multiplexing $ mvn clean install $ java -jar target/wootz-1.0-SNAPSHOT.jar src/main/resources/inception_v1.prototxt Saved file in location: inception_v1.py =================================================================================================== 2. To run inception_v1.prototxt with multiplexing $ mvn clean install $ java -jar target/wootz-1.0-SNAPSHOT.jar src/main/resources/inception_v1.prototxt true Saved file in location: inception_v1-multiplex.py =================================================================================================== 3. To run resnet50.prototxt with multiplexing $ mvn clean install $ java -jar target/wootz-1.0-SNAPSHOT.jar src/main/resources/resnet50.prototxt Saved file in location: resnet50.py =================================================================================================== 4. To run alexnet.prototxt with multiplexing $ mvn clean install $ java -jar target/wootz-1.0-SNAPSHOT.jar src/main/resources/alexnet.prototxt Saved file in location: alexnet.py ===================================================================================================
About
Java based source-to-source compiler that converts Deep Neural Network model from Caffe Prototxt to TensorFlow. It is heavily inspired by apcahe/incubator-mxnet.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published