job is failed: Error: NameError: name 'kmeans' is not defined

[HermannLizard]

运行pipcook主页的例子《识别图片中的前端组件》时报错：
job is failed: Error: NameError: name 'kmeans' is not defined
At:
./node_modules/@pipcook/plugins-pytorch-yolov5-model-train/tools/general.py(100): kmean_anchors
./node_modules/@pipcook/plugins-pytorch-yolov5-model-train/tools/general.py(623): check_anchors
./node_modules/@pipcook/plugins-pytorch-yolov5-model-train/init.py(141): train
./node_modules/@pipcook/plugins-pytorch-yolov5-model-train/init.py(289): main

at Function.value (/Users/hermannlizard/.pipcook/server/node_modules/@pipcook/boa/lib/index.js:204:25)
at Object.apply (/Users/hermannlizard/.pipcook/server/node_modules/@pipcook/boa/lib/index.js:401:26)
at /Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/loaders/python.js:8:24
at /Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/entry.js:142:32
at Generator.next (<anonymous>)
at /Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/entry.js:27:71
at new Promise (<anonymous>)
at __awaiter (/Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/entry.js:23:12)
at emitStart (/Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/entry.js:85:12)
at /Users/hermannlizard/.pipcook/server/node_modules/@pipcook/costa/dist/src/client/entry.js:202:13

后面还有：
job is failed: Error: NameError: name 'tqdm' is not defined
job is failed: TypeError: ModuleNotFoundError: No module named 'PIL'

[yorkie]

@HermannLizard @rickycao-qy could this be closed?

[HermannLizard]

@HermannLizard @rickycao-qy could this be closed?

OK. Everything is running for me now. Thx!
However, model train process is verrrrrrrry slow...
`
Starting training for 300 epochs...
⚠ run pipeline from pipeline dm6w7alu
⚠ Epoch gpu_mem box obj cls total targets img_size
⚠ Epoch: 0

⚠ 0/299 0G 0.07528 0.0932 0.01748 0.186 192 640
⚠ Current Best Map: [0.04981411]
⚠ Epoch: 1
`
Epoch: 0 took about 2h on my pc, and I have another 299 epochs coming…

My mac info:
MacBook Pro (16-inch, 2019)
2.6 GHz 六核Intel Core i7
16 GB 2667 MHz DDR4
AMD Radeon Pro 5300M 4 GB
Intel UHD Graphics 630 1536 MB

Is that normal for this PC taking this long to run trainings? Or did I have something wrong in my configuration, any suggestions? Thanks a lot

[FeelyChau]

@HermannLizard Generally speaking, using NVIDIA's GPU can speed up the training, but each epoch takes 2 hours, which is abnormal.