Blame - NnapiSupport.txt - ml/android-nn-driver - Gitiles

blob: c8f77e5f28bab006d0a5c3bba8a5de2796437e2b [file] [log] [blame]

surmeh01	49b9e10	2018-05-17 14:11:25 +0100	[diff] [blame]	1	------ ArmNN for Android NNAPI supported operations ------
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	2
Matthew Bentham	a3e23ca	2019-05-13 12:49:59 +0100	[diff] [blame^]	3	This release of ArmNN for Android supports use as a driver for the Android Neural Networks API. It implements the
				4	android.hardware.neuralnetworks@1.0 and android.hardware.neuralnetworks@1.1 HAL interfaces.
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	5
				6	For more information on the Android Neural Networks API, see https://developer.android.com/ndk/guides/neuralnetworks/index.html
				7
				8	For integration and usage documentation, please see README.md.
				9
				10	--- Support for Android Neural Networks HAL operations ---
				11
				12	The following AndroidNN operations are currently supported.
				13
				14	AndroidNN operator Tensor type supported
Matthew Bentham	378333d	2018-10-01 16:07:27 +0100	[diff] [blame]	15	ADD (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	16	AVERAGE_POOL_2D (FLOAT32,QUANT8_ASYMM)
Sadik Armagan	82fbeb3	2018-12-20 17:12:26 +0000	[diff] [blame]	17	BATCH_TO_SPACE_ND (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	18	CONCATENATION (FLOAT32)
surmeh01	49b9e10	2018-05-17 14:11:25 +0100	[diff] [blame]	19	CONV_2D (FLOAT32,QUANT8_ASYMM)
				20	DEPTHWISE_CONV_2D* (FLOAT32,QUANT8_ASYMM)
Ferran Balaguer	d04c043	2018-11-15 14:48:05 +0000	[diff] [blame]	21	DIV (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	22	FLOOR (FLOAT32)
Matthew Bentham	378333d	2018-10-01 16:07:27 +0100	[diff] [blame]	23	FULLY_CONNECTED (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	24	L2_NORMALIZATION (FLOAT32)
				25	L2_POOL_2D (FLOAT32)
				26	LOCAL_RESPONSE_NORMALIZATION (FLOAT32)
				27	LOGISTIC (FLOAT32,QUANT8_ASYMM)
Ferran Balaguer	d04c043	2018-11-15 14:48:05 +0000	[diff] [blame]	28	LSTM (FLOAT32)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	29	MAX_POOL_2D (FLOAT32,QUANT8_ASYMM)
Ferran Balaguer	d04c043	2018-11-15 14:48:05 +0000	[diff] [blame]	30	MEAN (FLOAT32,QUANT8_ASYMM)
Matthew Bentham	378333d	2018-10-01 16:07:27 +0100	[diff] [blame]	31	MUL (FLOAT32,QUANT8_ASYMM)
Ferran Balaguer	d04c043	2018-11-15 14:48:05 +0000	[diff] [blame]	32	PAD (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	33	RELU (FLOAT32,QUANT8_ASYMM)
				34	RELU1 (FLOAT32,QUANT8_ASYMM)
				35	RELU6 (FLOAT32,QUANT8_ASYMM)
				36	RESHAPE (FLOAT32,QUANT8_ASYMM)
				37	RESIZE_BILINEAR (FLOAT32)
				38	SOFTMAX (FLOAT32,QUANT8_ASYMM)
Sadik Armagan	82fbeb3	2018-12-20 17:12:26 +0000	[diff] [blame]	39	SPACE_TO_BATCH_ND (FLOAT32,QUANT8_ASYMM)
saoste01	b847148	2018-10-10 09:44:51 +0100	[diff] [blame]	40	SQUEEZE (FLOAT32,QUANT8_ASYMM)
Sadik Armagan	82fbeb3	2018-12-20 17:12:26 +0000	[diff] [blame]	41	STRIDED_SLICE (FLOAT32,QUANT8_ASYMM)
Ferran Balaguer	d04c043	2018-11-15 14:48:05 +0000	[diff] [blame]	42	SUB (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	43	TANH (FLOAT32)
saoste01	fe46315	2018-10-18 17:49:56 +0100	[diff] [blame]	44	TRANSPOSE (FLOAT32,QUANT8_ASYMM)
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	45
surmeh01	49b9e10	2018-05-17 14:11:25 +0100	[diff] [blame]	46	* Depthwise convolution only supports a value of 1 for the depth multiplier. In addition, the QUANT8_ASYMM version only supports 3x3 kernels.
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	47
				48	--- Unsupported operators ---
				49
telsoa01	ce3e84a	2018-08-31 09:31:35 +0100	[diff] [blame]	50	The following AndroidNN 1.0 operations are currently not supported.
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	51
				52	DEPTH_TO_SPACE
				53	DEQUANTIZE
				54	EMBEDDING_LOOKUP
				55	HASHTABLE_LOOKUP
				56	LSH_PROJECTION
telsoa01	5307bc1	2018-03-09 13:51:08 +0000	[diff] [blame]	57	RNN
				58	SPACE_TO_DEPTH
				59	SVDF
				60
				61	Where operations are not supported by the ArmNN Android NN Driver, the driver indicates this to the framework appropriately and the framework implements those operations using a CPU implementation.