nngen

nngen copied to clipboard

Consider taping out a test accelerator for no cost using open source tooling + PDK

1

Thanks to the funded provided by DARPA there is an increasingly capable, fully automated, RTL to GDS toolchain for creation ASICs called OpenROAD (https://theopenroadproject.org/). To help support the project, Google...

mithro

Request for a more detailed guide on how to implement generated hardware on FPGA

4

Hello everyone, As someone who has almost no background in digital electronics a more detailed guide (or perhaps a walkthrough using one of the example codes) on how to implement...

ChristiaanBoe

NNgen log ignores lazy_reshape, which is the output layer, and shows the previous operator as the output layer.

1

NNgen log ignores lazy_reshape, which is the output layer, and shows the previous operator as the output layer. ## the test case * /nngen/tests/matrix_reshape/test_matrix_reshape_int16.py ```py from __future__ import absolute_import from...

RyusukeYamano

function [nngen/basic_type.py:__Operator collect_numerics] is too busy

1

The function does not terminate in deep (& complex) networks because it re-explores the already explored ”args” when exploring the computational graph. https://github.com/NNgen/nngen/blob/2a30d83a040871f37a1d3bb17ac5e314d1b9d1b6/nngen/basic_types.py#L476-L484 ## This seems to work well. ```...

RyusukeYamano

test sample building error

2

Hello, I tried to run an example, but I got the following errors /nngen/tests/onnx_matrix_conv2d_linear/test_onnx_matrix_conv2d_linear_int32_3x3_stride1_3x3_stride1.py which is not the same as the test file we want to collect: Could you please...

BOPsemi

KeyError: 'MatMul'

1

Hi, after I generate the onnx file using a Lenet 5 pytorch model, running the code `dtypes = {} (outputs, placeholders, variables, constants, operators) = ng.from_onnx(onnx_filename, value_dtypes=dtypes, default_placeholder_dtype=act_dtype, default_variable_dtype=weight_dtype, default_constant_dtype=weight_dtype,...

qq757465678

KeyError, (CenterNet Pytorch)

7

### issue I've tried to load an onnx-model that I had created using a module, "torch.onnx.export", from a ".pth"(Pytorch) model. The following codes are for these implimentaitions. Some nodes in...

Hikaru-Furuta

I see that for the conversiom from ONNX to NNgen the usage of Mish layers is not supported. Can this please be added? Thank you in advance

ChristiaanBoe

Error while running nngen.from_onnx() Download the onnx file: https://github.com/onnx/models/blob/master/vision/body_analysis/arcface/model/arcfaceresnet100-8.onnx And run Python code: ```python import os import sys sys.path.insert(0, os.path.dirname(os.path.dirname( os.path.dirname(os.path.abspath(__file__))))) import nngen as ng act_dtype=ng.int16 weight_dtype=ng.int8 bias_dtype=ng.int32 scale_dtype=ng.int8 disable_fusion=False...

pbelecky

KeyError: 'Softmax'

2

Error while running nngen.from_onnx() Download the onnx file: https://github.com/onnx/models/blob/master/vision/classification/alexnet/model/bvlcalexnet-3.onnx And run Python code: ```python import os import sys sys.path.insert(0, os.path.dirname(os.path.dirname( os.path.dirname(os.path.abspath(__file__))))) import nngen as ng act_dtype=ng.int16 weight_dtype=ng.int8 bias_dtype=ng.int32 scale_dtype=ng.int8 disable_fusion=False...

pbelecky