hi, when I train vanilla nerf, there are countless threads, and sever is about to freeze. I found that this occurs due to https://github.com/google/nerfactor/blob/main/nerfactor/networks/embedder.py `freq_bands = 2. ** tf.linspace(0., log2_max_freq, n_freqs)`
hi, when I train vanilla nerf, there are countless threads, and sever is about to freeze.
I found that this occurs due to https://github.com/google/nerfactor/blob/main/nerfactor/networks/embedder.py
freq_bands = 2. ** tf.linspace(0., log2_max_freq, n_freqs)