-
Fyi
pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu121
Change device_map to cuda on text_model.py file.
**_Change cpu_**
```
class …
-
Hello after training Qlora I got produce checkpoint under
```
ll output/lora_vision_test/
adapter_config.json
adapter_model.safetensors
checkpoint-178/
config.json
non_lora_state_dict.bin
…
-
### bug描述 Describe the Bug
在 Jetpack 使用GPU 对PaddleOCR进行文字识别训练得到 warp-ctc [version 2] Error in get_workspace_size: execution failed。
如果使用CPU(Global.use_gpu=false)可以正常训练,但是使用GPU(Global.use_gpu=true)…
-
### bug描述 Describe the Bug
paddle-gpu2.6.0运行maskrcnn时出现ValueError: (InvalidArgument) multiclass_nms3(): argument (position 4) must be double, but got str错误:
![image](https://github.com/PaddlePaddle/…
-
```
from phi_3_vision_mlx import Agent
agent = Agent()
print(agent('Analyze this image and describe the architectural style:', 'https://images.metmuseum.org/CRDImages/rl/original/DP-19531-075.jpg…
-
### 🚀 The feature, motivation and pitch
phi-3.5 is a strong model for its size, including strong multi-image vision support. But vllm does not support the multi-image case.
https://github.com/vl…
-
### Feature request
Microsoft published its light-weight version of multi-modal model which also supports images. Is there any possibility to support this model in TGI?
### Motivation
Supporting mu…
-
**Is your feature request related to a problem? Please describe.**
[Phi-3](https://huggingface.co/collections/microsoft/phi-3-6626e15e9585a200d2d761e3) is a family of small language models and mult…
-
I use the following code in my PreProcessor under a Mirth Connect Channel:
// initialize a first_iteration flag.
// we don't want to add a carriage return
// before the MSH segment
var first_ite…
-
I use the following code in my PreProcessor under a Mirth Connect Channel:
// initialize a first_iteration flag.
// we don't want to add a carriage return
// before the MSH segment
var first_ite…