-
拉取了wespeaker最新的代码, 发现gpu占用率不能稳定, 经常为0, 猜测是cpu读取数据 太慢造成gpu空闲
batch size等dataloader配置都是默认值, 跑1000个batch差不多需要0.5h小时, 如果按照提供的配置来看,
在voxceleb数据集上训练150个epoch需要好几天
我的代码:
train_label = "data/vox2_…
-
Hello, I've followed a mix of the docker examples and the quicksetup description to setup up ot-recorder. It works (using MQTT but that's nor important here).
My question is about reverse geocoding…
sehe updated
5 months ago
-
- azure-storage-blob
- 12.12.0 (also tried 12.13.1 and 12.21.0)
- 20.04.6 LTS Ubuntu
- 3.10.14
**Describe the bug**
I have setup my `BlobServiceClient` and I am able to check that it exists, up…
-
### 软件环境
```Markdown
- paddlepaddle:
- paddlepaddle-gpu:
- paddlenlp:
paddle2onnx 1.1.0
paddlefsl 1.1.0
paddlenlp 2.7.2
paddlepaddle …
-
Computing eval metrics: 0%| | 0/86 [00:00
-
After some time running the genLMDB.py it stops. I have no idea what is happening
I hope someone can help me
I run the genLMDB.py and it works for a while until one of the following errors pops:
…
-
Traceback (most recent call last):
File "scripts/train_ted.py", line 179, in
main({'args': _args})
File "scripts/train_ted.py", line 133, in main
train_dataset = SpeechMotionDataset(a…
-
Thank you for sharing pretrained models and tutorials!
1. I noticed that when training with multi-GPUs, I need to create a metadata.npz file as mentioned in the **TRAIN.md** file, but when using OC…
-
### Description of the bug | 错误描述
Yesterday, the deployment of the CPU was completed.
Today, an attempt was made to deploy the GPU, but some problems were encountered.
### How to reproduce the bu…
-
I am trying to finetune the mobilenet_iter_73000.caffemodel with a subset of COCO.
I ran the normal Caffe VGG retraining with this same dataset and it runs, but then I realised I want MobileNet so I …