NNgen nngen issues - Githubissues

NNgen / nngen

NNgen: A Fully-Customizable Hardware Synthesis Compiler for Deep Neural Network

Apache License 2.0

339 stars 46 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

install issue

#67 sametyle opened 3 months ago
0
Error Generating model from ONNX

#66 kanndil opened 6 months ago
1
Sample code for ONNX doesn't work for mobilenet_v2

#65 MaKaRoIIIKa opened 12 months ago
0
supported FPGA

#64 Angelafat opened 1 year ago
0
fix verify rshift round

#63 CHENZIYANG0805 closed 1 year ago
0
Does nngen support operations involving sparse/dense matrices and vectors ?

#62 chenshih1 opened 1 year ago
0
Problems with the rshif_round

#61 CHENZIYANG0805 closed 1 year ago
0
Fix input_scale_factors

#60 hirayamy opened 1 year ago
0
IndexError: tuple index out of range in Gemm

#59 hirayamy closed 1 year ago
1
Implemented ONNX MatMul converter.

#58 hirayamy opened 1 year ago
0
Fix onnx/gemm.py to pass the tests/onnx_matrix_linear.

#57 hirayamy opened 1 year ago
0
added the logical not, and, and or operators

#56 sh-mug opened 1 year ago
0
implement tanh activation function

#55 sh-mug opened 1 year ago
0
v1.3.3 test faild: test_matrix_transpose_int16.py

#54 RyusukeYamano closed 2 years ago
1
v1.3.3 test faild: test_matrix_max_pool_int16_ksize7_stride7_global_par2.py

#53 RyusukeYamano closed 2 years ago
1
v1.3.3 test faild: test_matrix_matmul_int16_non_transposed_b.py

#52 RyusukeYamano closed 2 years ago
1
v1.3.3 test faild: test_matrix_avg_pool_int16_ksize7_stride7_global.py

#51 RyusukeYamano closed 2 years ago
1
v1.3.3 test faild: matrix_add_use_map_ram

#50 RyusukeYamano closed 2 years ago
1
fix *_imm bugs in verify (x_point is undefined)

#49 hashi0203 closed 1 year ago
0
n00b question: how to figure out latency?

#48 makslevental opened 2 years ago
0
setup.py is BROKEN

#47 KourFrost opened 2 years ago
1
Regression test (pytest) stucks issues found

#46 hanyong97 opened 2 years ago
0
Version constraints on the required ONNX modules are too strong.

#45 RyusukeYamano closed 2 years ago
3
Consider taping out a test accelerator for no cost using open source tooling + PDK

#44 mithro opened 2 years ago
1
Register is assinged in multiple always blocks

#43 garikyes closed 2 years ago
1
#verify : FAILED

#42 chen-yu-quan opened 2 years ago
4
Fixed low bit weight(binary, ternary, log) conv2d tests.

#41 kmichikura closed 2 years ago
0
Bug fix of concat operator with 3 or more inputs.

#40 kmichikura closed 2 years ago
0
NNgen log ignores lazy_reshape, which is the output layer, and shows the previous operator as the output layer.

#39 RyusukeYamano opened 3 years ago
1
function [nngen/basic_type.py:__Operator collect_numerics] is too busy

#38 RyusukeYamano closed 2 years ago
2
Fix cparam literal format

#37 RyusukeYamano closed 2 years ago
0
fix verify matmul

#36 RyusukeYamano closed 2 years ago
2
KeyError: 'MatMul'

#35 qq757465678 opened 3 years ago
1
Request for a more detailed guide on how to implement generated hardware on FPGA

#34 ChristiaanBoe opened 3 years ago
4
Bug fix for asymmetric_clip options in scaled_add and scaled_concat operators

#33 shtaxxx closed 3 years ago
1
KeyError, (CenterNet Pytorch)

#32 Hikaru-Furuta opened 3 years ago
7
Mish layer not supported

#31 ChristiaanBoe opened 3 years ago
0
Updated the behavior control of asymmetic_clip (imbalanced_clip)

#30 shtaxxx closed 3 years ago
1
KeyError: 'Softmax'

#29 pbelecky opened 4 years ago
2
KeyError: 'MatMul'

#28 pbelecky opened 4 years ago
0
KeyError: 'Dropout'

#27 pbelecky opened 4 years ago
0
Slice operator requires large RAM

#26 RyusukeYamano opened 4 years ago
0
畳み込みのMUX間のレジスタ削減

#25 RyusukeYamano closed 3 years ago
0
clipのunsigned対応

#24 x74ea6f closed 3 years ago
1
Verification environment

#23 oimasa5 opened 4 years ago
0
Feature imbalance clip

#22 RyusukeYamano closed 3 years ago
0
Feature register map

#21 RyusukeYamano closed 3 years ago
0
Fix rounding related to conv2d, leaky_relu and avg_pool

#20 RyusukeYamano closed 3 years ago
0
`Width of bit range is huge; vector of over 1billion bits: 0x40000000` occured in torchvision_onnx_resnet18 example

#19 in-die-nibelungen opened 4 years ago
2
feature add interrupt circit

#18 RyusukeYamano closed 4 years ago
0