-
### Expected behavior
Support of variable zero point in qnn batch_matmul
### Actual behavior
Quantized [model](https://huggingface.co/philschmid/quantized-distilbert-banking77?text=I+like+you.+I+…
-
app/src/main /AndroidManifest.xml中libcdsprpc.so依赖,属于无用依赖。
建议删除app/src/main /AndroidManifest.xml中libcdsprpc.so依赖,从而可以在非小米系列中运行
-
DDR size = 16GB
./main_qwen_npu -s 64 -c 1 -l 512
below is tail logs
`
Memory Usage: 8910 MB(19036) at: execute graph: 94
chunk:1 execute qnn graph 95
model.layers.23.self_attn.or_split …
-
执行quantize_torch_model.py后,pytorch模型export出的onnx模型并不是QNN模型,虽然export导出了一份json文件和onnx模型,但是有没有办法获得一个QNN量化后的int8模型(体积比原fp32模型小的)?
-
### Environment
- **Qiskit Machine Learning version**: 0.7.2
- **Python version**:
- **Operating system**:
### What is happening?
When the number of classical registers (measurements) is les…
-
### What is the expected enhancement?
VQC & QNN does not allow to implement Sklearn BaseEstimator to be inherited. Hence, we are unable to use Sklearn Gridsearch on VQC & QNN estimators for hyperpara…
-
### Describe the issue
The context cache
```
qnnOptions["qnn_context_cache_enable"] = "1";
qnnOptions["qnn_context_cache_path"] = contextCache;
```
only works if the graph can be executed in…
-
您好,
冒昧打扰您了。
我是一个AI领域的小白,对贵公司的这个开源项目很感兴趣(打算移植到个人AI学习/研究开源项目中)。
昨天半夜下载了这个代码。今天上午在市场上购买的小米14上花了半天(4个小时)测试,目前遇到如下问题:
(1)总是提示无法加载模型
看了下c++代码,发现找不到模型文件与高通库(代码里dump出来的路径如果没有root权限,无法访问)。
ad…
-
我成功编译出apk ,并将模型放置于src/main/asserts/stable_diffusion/qnn_model_8550下,但是我运行的apk的时候出现模型加载失败?请问这是什么原因导致的。我已经严格按照readme设置了。
-
### Describe the issue
QNN's implementation of MatMul is much slower than the equivalent FullyConnected implementation.
The QNN delegate should insert reshapes around 3D/4D MatMul to squish the in…
kory updated
4 months ago