-
https://arxiv.org/pdf/2401.14351.pdf
-
### System Info
- `transformers` version: 4.36.2
- Platform: Linux-5.15.0-91-generic-x86_64-with-glibc2.31
- Python version: 3.11.5
- Huggingface_hub version: 0.20.1
- Safetensors version: 0.4.…
-
问题描述: 可以通过localhost:9997加载出web ui,但是在注册本地模型或者加载内置模型(包括LLM和embedding),下载完成后运行时都会报这个错
`RuntimeError: [address=0.0.0.0:36297, pid=1139] Failed to import transformers.generation.utils because of the fo…
-
### 1. System information
- OS Platform and Distribution: iOS / Android
### 2. Code
I download the [MiDaS ONNX model](https://github.com/isl-org/MiDaS/releases/download/v2_1/model-small.onnx)…
-
### Your current environment
I am currently using a T4 instance on Google Colaboratory.
```
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used…
-
https://arxiv.org/pdf/2302.11665.pdf
-
I have a 7900XT and would definitely love to have ROCm support. It seems like it might be coming with https://github.com/jmorganca/ollama/pull/667?
I couldn't find a dedicated issue for this so I'm…
-
### 🐛 Describe the bug
`ExecutorchCallDelegateHigherOrderVariable` will attempt to retrieve the real values as inputs to the executorch module, and then deep copy fake the result into new fake valu…
-
For context this code is run on a tenstorrent grayskull e150.
If I run this
```python
# try the forward pass on the tt-card
module = pybuda.PyTorchModule("direct_pt", myNet())
tt0…
-
**MY QUESTION**
Training a model with FLOAT-32 and converts into a saved model. Then I use **'pos_training_quantized=True '** to generate .tflite model (int8).
according to official website: http…