hanhx

> Currently, we only support np.float32 type. np.random.uniform will generate np.float64 data. Try to use np.random.uniform(low=0.5, high=13.3, size=(n,)).astype(np.float32) Passed, thanks

yep, It takes so long to complete the training index I've tried on the experiment with the dataset (1024 * 1024 * 5, 128) -- 5.242.880 vectors of 128 dim....

> > yep, It takes so long to complete the training index > > I've tried on the experiment with the dataset (1024 * 1024 * 5, 128) -- 5.242.880...

solved