Ansh Rai 0ae2c5391c mlearning/tflite-micro: Register DEPTHWISE_CONV_2D in tflm_tool resolver. ... MicroMutableOpResolver<8> only registered 8 ops, missing DEPTHWISE_CONV_2D, required by any depthwise-separable CNN (MobileNet-style, DS-CNN keyword-spotting models). The kernel already exists upstream (Register_DEPTHWISE_CONV_2D_INT8() in tensorflow/lite/micro/kernels/depthwise_conv.h); this wires it into the resolver and bumps the template size to <9>. Verified with micro_speech_quantized.tflite on sim:tflm. Before: Didn't find op for builtin opcode 'DEPTHWISE_CONV_2D'. After: RESHAPE/DEPTHWISE_CONV_2D/FULLY_CONNECTED/SOFTMAX all execute. Signed-off-by: Ansh Rai <anshrai331@gmail.com>		2026-09-07 14:07:09 +02:00
..
operators/neon	mlearning: migrate to SPDX identifier	2024-12-21 13:53:10 +08:00
.gitignore	fix: git status/git status --ignore normalization check problem	2024-10-18 09:40:17 +08:00
0001-dequantize-int8.patch	ml: tflm dequantize optimization patch	2024-10-18 09:40:17 +08:00
0002-quantize-int8.patch	ml: tflm quantize optimization patch	2024-10-18 09:40:17 +08:00
0003-mean-int8.patch	ml: tflm mean optimization patch	2024-10-18 09:40:17 +08:00
0004-tflite-add-extern-C-to-main-function-to-avoid-c-mang.patch	tflm: add tflm hello world example into nuttx build	2024-12-02 13:33:55 +08:00
CMakeLists.txt	cmake: Use NUTTX(_DIR/_BINARY_DIR) instead of CMAKE(_SOURCE_DIR/BINARY_DIR)	2026-08-09 10:43:01 -03:00
Kconfig	mlearning/tflite-micro: fixed the header file	2026-08-11 15:38:46 -04:00
Make.defs	mlearning: migrate to SPDX identifier	2024-12-21 13:53:10 +08:00
Makefile	mlearning: migrate to SPDX identifier	2024-12-21 13:53:10 +08:00
tflite-micro.patch	delete the neon conv2D	2024-10-18 09:40:17 +08:00
tflm_syslog.cc	mlearning: migrate to SPDX identifier	2024-12-21 13:53:10 +08:00
tflm_tool.cc	mlearning/tflite-micro: Register DEPTHWISE_CONV_2D in tflm_tool resolver.	2026-09-07 14:07:09 +02:00