Commit Graph

8 Commits

Author SHA1 Message Date
Zhen Kong
1a88104160
Remove module instance from hashmap in wasi_nn_destroy (#2613)
When destroying wasi-nn context, module instance should be also
removed from hashmap to avoid memory leak.
2023-10-03 08:33:11 +08:00
tonibofarull
b45d014112
wasi-nn: Improve TPU support (#2447)
1. Allow TPU and GPU support at the same time.
2. Add Dockerfile to run example with [Coral USB](https://coral.ai/products/accelerator/).
2023-08-14 20:03:56 +08:00
tonibofarull
0b0af1b3df
wasi-nn: Support uint8 quantized networks (#2433)
Support (non-full) uint8 quantized networks.
Inputs and outputs are still required to be `float`. The (de)quantization is done internally by wasi-nn.

Example generated from `quantized_model.py`:
![Screenshot from 2023-08-07 17-57-05](https://github.com/bytecodealliance/wasm-micro-runtime/assets/80318361/91f12ff6-870c-427a-b1dc-e307f7d1f5ee)

Visualization with [netron](https://netron.app/).
2023-08-11 07:55:40 +08:00
tonibofarull
ab96e01f5e
wasi-nn: Add support of wasi-nn as shared lib (#2310)
## Context

Currently, WAMR supports compiling iwasm with flag `WAMR_BUILD_WASI_NN`.
However, there are scenarios where the user might prefer having it as a shared library.

## Proposed Changes

Decouple wasi-nn context management by internally managing the context given
a module instance reference.
2023-06-27 18:18:26 +08:00
ayakoakasaka
89be5622a5
wasi-nn: Add external delegation to support several NPU/GPU (#2162)
Add VX delegation as an external delegation of TFLite, so that several NPU/GPU
(from VeriSilicon, NXP, Amlogic) can be controlled via WASI-NN.

Test Code can work with the X86 simulator.
2023-05-05 16:29:36 +08:00
tonibofarull
a15a731e12
wasi-nn: Support multiple TFLite models (#2002)
Remove restrictions:
- Only 1 WASM app at a time
- Only 1 model at a time
   - `graph` and `graph-execution-context` are ignored

Refer to previous document:
e8d718096d/core/iwasm/libraries/wasi-nn/README.md
2023-03-08 15:54:06 +08:00
tonibofarull
1614ce12fa
wasi-nn: Enable GPU support (#1922)
- Split logic in several dockers
  - runtime: wasi-nn-cpu and wasi-nn- Nvidia-gpu.
  - compilation: wasi-nn-compile. Prepare the testing wasm and generates the TFLites.
- Implement GPU support for TFLite with Opencl.
2023-02-02 08:09:46 +08:00
tonibofarull
9eed6686df
Refactor WASI-NN to simplify the support for multiple frameworks (#1834)
- Reorganize the library structure
- Use the latest version of `wasi-nn` wit (Oct 25, 2022):
    0f77c48ec1/wasi-nn.wit.md
- Split logic that converts WASM structs to native structs in a separate file
- Simplify addition of new frameworks
2023-01-25 18:32:40 +08:00