FakeHeteroDataset generates graphs with higher mean node degrees than avg_degree

[MatthijsdeJ]

🐛 Describe the bug

Graphs generated with FakeHeteroDataset have a higher mean node degree than specified with the avg_degree parameter. If averaged over sufficiently many graphs, I would expect the mean node degree to converge to the avg_degree parameter. However, this is not true, as inestigated with the MRE:

import torch_geometric as tg
import numpy as np
import matplotlib.pyplot as plt

MAX_AVR_DEGREE = 15
SAMPLES_PER_STEP = 100

mean_degrees = {}
for i in np.arange(1,MAX_AVR_DEGREE):
    mean_degrees[i] = []

    for _ in range(SAMPLES_PER_STEP):

        data = tg.datasets.FakeDataset(avg_num_nodes=32, avg_degree=i)
        mean_degree = tg.utils.degree(data.edge_index[0]).mean() # data.edge_index.shape[1] / data.x.shape[0]
        mean_degrees[i].append(mean_degree)
    
mean_degrees = {k:np.mean(v) for k, v in mean_degrees.items()}
plt.scatter(np.arange(1, MAX_AVR_DEGREE), mean_degrees.values())
plt.xlabel('avg_degree')
plt.ylabel('Mean (over samples) mean (over nodes) degree')

Resulting plot:

As visible the specified avg_degree (x axis) and computed mean degree (y axis) diverge. This seems like unintended behavior.

Versions

Collecting environment information...
PyTorch version: 2.5.1
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: macOS 14.4.1 (arm64)
GCC version: Could not collect
Clang version: 15.0.0 (clang-1500.3.9.4)
CMake version: Could not collect
Libc version: N/A

Python version: 3.11.0 | packaged by conda-forge | (main, Jan 14 2023, 12:26:40) [Clang 14.0.6 ] (64-bit runtime)
Python platform: macOS-14.4.1-arm64-arm-64bit
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Apple M1

Versions of relevant libraries:
[pip3] flake8==7.0.0
[pip3] mypy==1.8.0
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.26.4
[pip3] numpydoc==1.5.0
[pip3] torch==2.5.1
[pip3] torch-geometric==2.6.1
[conda] numpy 1.26.4 py311he598dae_0
[conda] numpy-base 1.26.4 py311hfbfe69c_0
[conda] numpydoc 1.5.0 py311hca03da5_0
[conda] torch 2.5.1 pypi_0 pypi
[conda] torch-geometric 2.6.1 pypi_0 pypi