vespa - An engine for low-latency computation over large data sets

	Commit message (Collapse)	Author	Age	Files	Lines
*	Remove Java dependencies to tensorflow	Lester Solbakken	2021-10-04	48	-102382/+0
\|
*	Add/corect copyright headers	Jon Bratseth	2020-01-03	2	-0/+2
\|
*	Initial conversion of TF to ONNX for testing	Lester Solbakken	2019-12-12	3	-0/+5039
\|
*	Add test for tf2onnx conversion and evaluation	Lester Solbakken	2019-11-22	17	-0/+23837
\|
*	Rename constant dimensions	Jon Bratseth	2019-07-28	4	-0/+6561
\|
*	Output the intermediate graph	Jon Bratseth	2019-06-30	1	-0/+1318
\|
*	Remove outdated comment	Lester Solbakken	2019-06-19	1	-3/+1
\|
*	Add TensorFlow softmax operation support	Lester Solbakken	2019-06-19	4	-0/+2030
\|
*	Separate model integration into a separate module	Jon Bratseth	2018-11-21	19	-0/+63597
	This allows us to access model importers (such as TensorFlow) in config models without loading one instance per config model instance, which is not possible with TensorFlow because it depends on JNI code.