InfProbSciX

import os, random
import torch, torchaudio

import numpy as np
import pandas as pd
from tqdm import tqdm
from scipy.stats import spearmanr

SUBSET_DATA = True
DATA_PARENT_DIR = '../ProbMOS/'

InfProbSciX

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import fetch_openml
from sklearn.random_projection import GaussianRandomProjection
from sklearn.neighbors import NearestNeighbors
from scipy.stats import spearmanr
from tqdm import tqdm

np.random.seed(42)

InfProbSciX

import os, torch, fairseq, random, numpy as np
device = 'cuda'

model, _, _ = fairseq.checkpoint_utils.load_model_ensemble_and_task(['wav2vec_large.pt'])
model = model[0].eval().to(device)
model.wav2vec_predictions.infonce = False

torch.manual_seed(42); np.random.seed(42); random.seed(42)
a = torch.nn.Parameter(torch.randn(1, 80000).to(device))

InfProbSciX

import torch
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from cmdstanpy import CmdStanModel
from sklearn.decomposition import PCA
from sklearn.neighbors import NearestNeighbors
from torchvision.transforms.functional import rotate

InfProbSciX

library(mgcv) # version >= 1.8-34
library(data.table)

set.seed(42)

n = 200; k = 30; bs_dim = 15; l_scale = 2
data = data.table(x = seq(-1, 1, l=n))
data[, y := x^2 + rnorm(n, sd=0.25)]

model = gam(y ~ s(x, bs="gp", k=bs_dim, m=c(-4, l_scale),