[INFO] 正在启动WebUI, 请稍等...
[INFO] Starting WebUI, please wait...
[INFO] 若启动失败, 请尝试删除软件根目录下data文件夹或以管理员身份运行
[INFO] If launch failed, try to delete the data folder under the root directory or run as administrator
[注意] WebUI运行过程中请勿关闭此窗口!
[NOTE] Do not close this window while WebUI is running!
[INFO] 设备信息:0: NVIDIA GeForce GTX 1060 6GB
Running on local URL: http://127.0.0.1:7860
To create a public link, set share=True in launch().
Use command: .\workenv\python.exe msst_inference.py --model_type bs_roformer --config_path "configs/viperx/model_bs_roformer_ep_368_sdr_12.9628.yaml" --start_check_point "pretrain/vocal_models\model_bs_roformer_ep_368_sdr_12.9628.ckpt" --input_folder "temp" --store_dir "C:/Users" --device_ids 0
23:37:47.240 [INFO] msst_inference - Using device: cuda:0
GPU Compute Capability below 8.0, using math or mem efficient attention if input tensor is on cuda
23:37:48.531 [INFO] msst_inference - Start from checkpoint: pretrain/vocal_models\model_bs_roformer_ep_368_sdr_12.9628.ckpt
23:37:49.390 [INFO] msst_inference - Instruments: ['vocals', 'instrumental']
23:37:49.588 [INFO] msst_inference - Total files found: 1
日志信息如下 似乎是 GPU计算能力低于8.0,无法使用GPU加速, 我使用的是1066 算力6.1 已更新最新版驱动,不确定我这个情况是否为个例 或是我使用方式有问题 只尝试了model_bs_roformer_ep_368_sdr_12.9628模型 其它未知 附上一张算力表 文档中似乎没提过对算力的要求 可以完善一下
[INFO] 正在启动WebUI, 请稍等... [INFO] Starting WebUI, please wait... [INFO] 若启动失败, 请尝试删除软件根目录下data文件夹或以管理员身份运行 [INFO] If launch failed, try to delete the data folder under the root directory or run as administrator [注意] WebUI运行过程中请勿关闭此窗口! [NOTE] Do not close this window while WebUI is running! [INFO] 设备信息:0: NVIDIA GeForce GTX 1060 6GB Running on local URL: http://127.0.0.1:7860
To create a public link, set
share=True
inlaunch()
. Use command: .\workenv\python.exe msst_inference.py --model_type bs_roformer --config_path "configs/viperx/model_bs_roformer_ep_368_sdr_12.9628.yaml" --start_check_point "pretrain/vocal_models\model_bs_roformer_ep_368_sdr_12.9628.ckpt" --input_folder "temp" --store_dir "C:/Users" --device_ids 0 23:37:47.240 [INFO] msst_inference - Using device: cuda:0 GPU Compute Capability below 8.0, using math or mem efficient attention if input tensor is on cuda 23:37:48.531 [INFO] msst_inference - Start from checkpoint: pretrain/vocal_models\model_bs_roformer_ep_368_sdr_12.9628.ckpt 23:37:49.390 [INFO] msst_inference - Instruments: ['vocals', 'instrumental'] 23:37:49.588 [INFO] msst_inference - Total files found: 1