blueoil icon indicating copy to clipboard operation
blueoil copied to clipboard

Bring Deep Learning to small devices

Results 100 blueoil issues
Sort by recently updated
recently updated
newest added

## Motivation and Context support sigmoid function. It is related to #163 ## Description Add operator sigmoid. Add sigmoid function generator. Define sigmoid function. Add header to use sigmoid function....

enhancement

Work in progress PR. try PACT quantization https://arxiv.org/abs/1805.06085 ## Motivation and Context ## Description ## How has this been tested? ## Screenshots (if appropriate): ## Types of changes - [...

CI: auto-run

## Motivation and Context Previously, there were not config file provided for object detection of the experimental parameter tuner. This PR provides it. ## Description - This PR includes the...

CI: auto-run

## Motivation and Context ## Description DL model for optical flow estimation ### How do you use it? [NOTE] Please run them on high end computer with GPU. #### Build...

CI: auto-run

In current implementation of converter, parameters (shape, dtype, ... or dimension) of tensor are member of `Operator` class, but `Split` operator can have multiple outputs. Because of that implementation, `Split`...

This is to fix the bug described in issue https://github.com/blue-oil/blueoil/issues/1076 The error occurs because when running convert, the [_saved_config_file_path](https://github.com/blue-oil/blueoil/blob/f57ccf6af2b5fd8210ba255a6ce46e868d9c2677/blueoil/utils/config.py#L64) function will retrieve the `config.py` file, and it will try to...

CI: auto-run

Now we have two configurations for buffers between CPU and FPGA. One is a configuration which use `/dev/mem` for all buffers (threshold/kernel/input/output). Another is a configuration which threshold/kernel use `/dev/mem`...

https://github.com/blue-oil/blueoil/blob/107c4fc71685159e6bfe077b263ac441d126b602/blueoil/converter/templates/include/dlk_test.h#L41-L57 In this above source, absolute error is used in comparing ground truth values and calculated values for elf binaries like `lm_x86.elf`. This comparison with absolute error works well when...

Now runtime only support one inference on FPGA. But sometimes running two models is useful. We might need the feature for that.

enhancement