SurvivalPFN

SurvivalPFN is a PyTorch checkpoint for survival / time-to-event modeling. This model repository currently hosts the trained checkpoint artifact; the code repository is being prepared at rgklab/SurvivalPFN.

Files

• survivalpfn_v0.1.pt: PyTorch checkpoint containing model weights and a sanitized model architecture config.
• checkpoint_info.json: Machine-readable summary of the uploaded checkpoint artifact.

Checkpoint Details

• Format: PyTorch zip checkpoint (.pt)
• Size: 75,435,990 bytes (~72 MiB)
• SHA-256: a7b844809ad2dc5a675384704f7b6d85dfb88e96b82031c4542ce90a4fd439bb
• Top-level checkpoint keys: model_state_dict, model_config
• Model state: 151 tensors, 18,844,032 tensor values
• Model config: sanitized architecture fields only

Loading

The full model definition will be provided in the SurvivalPFN code repository. Once available, instantiate the model from that codebase using checkpoint["model_config"] and load checkpoint["model_state_dict"].

import torch

checkpoint = torch.load(
    "survivalpfn_v0.1.pt",
    map_location="cpu",
    weights_only=False,  # Use only for checkpoints from sources you trust.
)

state_dict = checkpoint["model_state_dict"]
model_config = checkpoint["model_config"]

Intended Use

This checkpoint is intended for research use with the SurvivalPFN implementation. It should not be used for clinical or other high-stakes decision making without independent validation in the target setting.