Deep Neural Networks for Survival Analysis using R torch
survdnn implements neural network-based models for right-censored
survival analysis using the native torch backend in R. It supports
multiple loss functions including Cox partial likelihood, L2-penalized
Cox, Accelerated Failure Time (AFT) objectives, as well as
time-dependent extension such as Cox-Time. The package provides a
formula interface, supports model evaluation using time-dependent
metrics (C-index, Brier score, IBS), cross-validation, and
hyperparameter tuning.
A methodological paper describing the design, implementation, and
evaluation of survdnn is currently under review at The R Journal.
-
Formula interface for
Surv() ~ .models -
Modular neural architectures: configurable layers, activations, optimizers, and losses
-
Built-in survival loss functions:
"cox": Cox partial likelihood"cox_l2": penalized Cox"aft": Accelerated Failure Time"coxtime": deep time-dependent Cox
-
Evaluation: C-index, Brier score, IBS
-
Model selection with
cv_survdnn()andtune_survdnn() -
Prediction of survival curves via
predict()andplot()
# Install from CRAN
install.packages("survdnn")
# Install from GitHub
install.packages("remotes")
remotes::install_github("ielbadisy/survdnn")
# Or clone and install locally
git clone https://github.com/ielbadisy/survdnn.git
setwd("survdnn")
devtools::install()library(survdnn)
library(survival, quietly = TRUE)
library(ggplot2)
veteran <- survival::veteran
mod <- survdnn(
Surv(time, status) ~ age + karno + celltype,
data = veteran,
hidden = c(32, 16),
epochs = 300,
loss = "cox",
verbose = TRUE
)
summary(mod)##
## Formula:
## Surv(time, status) ~ age + karno + celltype
## <environment: 0x6011c835fbe0>
##
## Model architecture:
## Hidden layers: 32 : 16
## Activation: relu
## Dropout: 0.3
## Batch norm: TRUE
## Final loss: 3.841008
##
## Training summary:
## Epochs: 300
## Learning rate: 1e-04
## Loss function: cox
## Optimizer: adam
## Device: cpu
## CPU threads: default
## NA action: omit
##
## Data summary:
## Observations (used/total): 137 / 137
## Predictors (5): age, karno, celltypesmallcell, celltypeadeno, celltypelarge
## Time range: [ 1, 999 ]
## Events / censored: 128 / 9
## Event rate: 93.4%
## Predictors standardized: yes
plot(mod, group_by = "celltype", times = 1:300)- Cox partial likelihood
mod1 <- survdnn(
Surv(time, status) ~ age + karno,
data = veteran,
loss = "cox",
epochs = 300
)## [survdnn::fit] start: n=137 p=2 loss=cox optimizer=adam epochs=300 device=cpu
## [survdnn::fit] epoch 50/300 loss=3.989828
## [survdnn::fit] epoch 100/300 loss=3.942912
## [survdnn::fit] epoch 150/300 loss=3.890888
## [survdnn::fit] epoch 200/300 loss=3.869117
## [survdnn::fit] epoch 250/300 loss=3.866454
## [survdnn::fit] epoch 300/300 loss=3.870147
## [survdnn::fit] done: epochs_run=300 final_loss=3.870147
- Accelerated Failure Time
mod2 <- survdnn(
Surv(time, status) ~ age + karno,
data = veteran,
loss = "aft",
epochs = 300
)## [survdnn::fit] start: n=137 p=2 loss=aft optimizer=adam epochs=300 device=cpu
## [survdnn::fit] epoch 50/300 loss=4.628552
## [survdnn::fit] epoch 100/300 loss=4.641005
## [survdnn::fit] epoch 150/300 loss=4.563893
## [survdnn::fit] epoch 200/300 loss=4.507134
## [survdnn::fit] epoch 250/300 loss=4.501204
## [survdnn::fit] epoch 300/300 loss=4.494009
## [survdnn::fit] done: epochs_run=300 final_loss=4.494009
- Coxtime
mod3 <- survdnn(
Surv(time, status) ~ age + karno,
data = veteran,
loss = "coxtime",
epochs = 300
)## [survdnn::fit] start: n=137 p=2 loss=coxtime optimizer=adam epochs=300 device=cpu
## [survdnn::fit] epoch 50/300 loss=3.946552
## [survdnn::fit] epoch 100/300 loss=3.955686
## [survdnn::fit] epoch 150/300 loss=3.944864
## [survdnn::fit] epoch 200/300 loss=3.930768
## [survdnn::fit] epoch 250/300 loss=3.838923
## [survdnn::fit] epoch 300/300 loss=3.889966
## [survdnn::fit] done: epochs_run=300 final_loss=3.889966
cv_results <- cv_survdnn(
Surv(time, status) ~ age + karno + celltype,
data = veteran,
times = c(600),
metrics = c("cindex", "ibs"),
folds = 3,
hidden = c(16, 8),
loss = "cox",
epochs = 300
)
print(cv_results)grid <- list(
hidden = list(c(16), c(32, 16)),
lr = c(1e-3),
activation = c("relu"),
epochs = c(100, 300),
loss = c("cox", "aft", "coxtime")
)
tune_res <- tune_survdnn(
formula = Surv(time, status) ~ age + karno + celltype,
data = veteran,
times = c(90, 300),
metrics = "cindex",
param_grid = grid,
folds = 3,
refit = FALSE,
return = "summary"
)
print(tune_res)tune_survdnn() can be used also to automatically refit the
best-performing model on the full dataset. This behavior is controlled
by the refit and return arguments. For example:
best_model <- tune_survdnn(
formula = Surv(time, status) ~ age + karno + celltype,
data = veteran,
times = c(90, 300),
metrics = "cindex",
param_grid = grid,
folds = 3,
refit = TRUE,
return = "best_model"
)In this mode, cross-validation is used to select the optimal
hyperparameter configuration, after which the selected model is refitted
on the full dataset. The function then returns a fitted object of class
"survdnn".
The resulting model can be used directly for prediction visualization, and evaluation:
summary(best_model)
plot(best_model, times = 1:300)
predict(best_model, veteran, type = "risk", times = 180)This makes tune_survdnn() suitable for end-to-end workflows, combining
model selection and final model fitting.
plot(mod1, group_by = "celltype", times = 1:300)
plot(mod1, group_by = "celltype", times = 1:300, plot_mean_only = TRUE)
help(package = "survdnn")
?survdnn
?tune_survdnn
?cv_survdnn
?plot.survdnn# run all tests
devtools::test()By default, {torch} initializes model weights and shuffles minibatches
using random draws, so results may differ across runs. Unlike
set.seed(), which only controls R’s random number generator, {torch}
relies on its own RNG implemented in C++ (and CUDA when using GPUs).
To ensure reproducibility, random seeds must therefore be set at the Torch level as well.
survdnn provides built-in control of randomness to guarantee
reproducible results across runs. The main fitting function,
survdnn(), exposes a dedicated .seed argument:
mod <- survdnn(
Surv(time, status) ~ age + karno + celltype,
data = veteran,
epochs = 300,
.seed = 123
)When .seed is provided, survdnn() internally synchronizes both R and
Torch random number generators via survdnn_set_seed(), ensuring
reproducible:
-
weight initialization
-
dropout behavior
-
minibatch ordering
-
loss trajectories
If .seed = NULL (the default), randomness is left uncontrolled and
results may vary between runs.
For full reproducibility in cross-validation or hyperparameter tuning,
the same .seed mechanism is propagated internally by cv_survdnn()
and tune_survdnn(), ensuring consistent data splits, model
initialization, and optimization paths across repetitions.
survdnn relies on the {torch} backend for numerical computation. The
number of CPU cores (threads) used during training, prediction, and
evaluation is controlled globally by Torch.
By default, Torch automatically configures its CPU thread pools based on the available system resources, unless explicitly overridden by the user using:
torch::torch_set_num_threads(4)You can also set this directly from survdnn APIs with .threads:
mod <- survdnn(
Surv(time, status) ~ age + karno + celltype,
data = survival::veteran,
.threads = 4
)## [survdnn::fit] start: n=137 p=5 loss=cox optimizer=adam epochs=300 device=cpu
## [survdnn::fit] cpu_threads=4
## [survdnn::fit] epoch 50/300 loss=3.962863
## [survdnn::fit] epoch 100/300 loss=3.952089
## [survdnn::fit] epoch 150/300 loss=3.896662
## [survdnn::fit] epoch 200/300 loss=3.834132
## [survdnn::fit] epoch 250/300 loss=3.866578
## [survdnn::fit] epoch 300/300 loss=3.821199
## [survdnn::fit] done: epochs_run=300 final_loss=3.821199
The same .threads argument is available in cv_survdnn() and
tune_survdnn().
This setting affects:
-
model training
-
prediction
-
evaluation metrics
-
cross-validation and hyperparameter tuning
GPU acceleration can be enabled by setting .device = "cuda" when
calling survdnn() (cv_survdnn() and tune_survdnn() too).
The survdnn R package is available on
CRAN or
github
Contributions, issues, and feature requests are welcome!
Open an issue or submit a pull request.
MIT License © 2025 Imad EL BADISY