tar_target_raw() is just like tar_target() except
it avoids non-standard evaluation for the arguments: name
is a character string, command and pattern are language objects,
and there is no tidy_eval argument. Use tar_target_raw()
instead of tar_target() if you are creating entire batches
of targets programmatically (metaprogramming, static branching).
tar_target_raw(
name,
command,
pattern = NULL,
packages = targets::tar_option_get("packages"),
library = targets::tar_option_get("library"),
deps = NULL,
string = NULL,
format = targets::tar_option_get("format"),
iteration = targets::tar_option_get("iteration"),
error = targets::tar_option_get("error"),
memory = targets::tar_option_get("memory"),
deployment = targets::tar_option_get("deployment"),
priority = targets::tar_option_get("priority"),
resources = targets::tar_option_get("resources"),
storage = targets::tar_option_get("storage"),
retrieval = targets::tar_option_get("retrieval"),
cue = targets::tar_option_get("cue")
)Character of length 1, name of the target.
Similar to the command argument of tar_target() except
the object must already be an expression instead of
informally quoted code.
base::expression() and base::quote() can produce such objects.
Similar to the pattern argument of tar_target()
except the object must already be an expression instead of
informally quoted code.
base::expression() and base::quote() can produce such objects.
Character vector of packages to load right before
the target builds. Use tar_option_set() to set packages
globally for all subsequent targets you define.
Character vector of library paths to try
when loading packages.
Optional character vector of the adjacent upstream
dependencies of the target, including targets and global objects.
If NULL, dependencies are resolved automatically as usual.
Optional string representation of the command.
Internally, the string gets hashed to check if the command changed
since last run, which helps targets decide whether the
target is up to date. External interfaces can take control of
string to ignore changes in certain parts of the command.
If NULL, the strings is just deparsed from command (default).
Optional storage format for the target's return value.
With the exception of format = "file", each target
gets a file in _targets/objects, and each format is a different
way to save and load this file.
Possible formats:
"rds": Default, uses saveRDS() and readRDS(). Should work for
most objects, but slow.
"qs": Uses qs::qsave() and qs::qread(). Should work for
most objects, much faster than "rds". Optionally set the
preset for qsave() through the resources argument, e.g.
tar_target(..., resources = list(preset = "archive")).
"fst": Uses fst::write_fst() and fst::read_fst().
Much faster than "rds", but the value must be
a data frame. Optionally set the compression level for
fst::write_fst() through the resources argument, e.g.
tar_target(..., resources = list(compress = 100)).
"fst_dt": Same as "fst", but the value is a data.table.
Optionally set the compression level the same way as for "fst".
"fst_tbl": Same as "fst", but the value is a tibble.
Optionally set the compression level the same way as for "fst".
"keras": Uses keras::save_model_hdf5() and
keras::load_model_hdf5(). The value must be a Keras model.
"torch": Uses torch::torch_save() and torch::torch_load().
The value must be an object from the torch package
such as a tensor or neural network module.
"file": A dynamic file. To use this format,
the target needs to manually identify or save some data
and return a character vector of paths
to the data. Those paths must point to files or directories,
and they must not contain characters | or *.
Then, targets automatically checks those files and cues the
appropriate build decisions if those files are out of date.
"url": A dynamic input URL. It works like format = "file"
except the return value of the target is a URL that already exists
and serves as input data for downstream targets. Optionally
supply a custom curl handle through the resources argument, e.g.
tar_target(..., resources = list(handle = curl::new_handle())).
The data file at the URL needs to have an ETag or a Last-Modified
time stamp, or else the target will throw an error because
it cannot track the data. Also, use extreme caution when
trying to use format = "url" to track uploads. You must be absolutely
certain the ETag and Last-Modified time stamp are fully updated
and available by the time the target's command finishes running.
targets makes no attempt to wait for the web server.
"aws_rds", "aws_qs", "aws_fst", "aws_fst_dt",
"aws_fst_tbl", "aws_keras": AWS-powered versions of the
respective formats "rds", "qs", etc. The only difference
is that the data file is uploaded to the AWS S3 bucket
you supply to resources. See the cloud computing chapter
of the manual for details.
"aws_file": arbitrary dynamic files on AWS S3. The target
should return a path to a temporary local file, then
targets will automatically upload this file to an S3
bucket and track it for you. Unlike format = "file",
format = "aws_file" can only handle one single file,
and that file must not be a directory.
tar_read() and downstream targets
download the file to _targets/scratch/ locally and return the path.
_targets/scratch/ gets deleted at the end of tar_make().
Requires the same resources and other configuration details
as the other AWS-powered formats. See the cloud computing
chapter of the manual for details.
Character of length 1, name of the iteration mode of the target. Choices:
"vector": branching happens with vectors::vec_slice() and
aggregation happens with vctrs::vec_c().
"list", branching happens with [[]] and aggregation happens with
list().
"group": dplyr::group_by()-like functionality to branch over
subsets of a data frame. The target's return value must be a data
frame with a special tar_group column of consecutive integers
from 1 through the number of groups. Each integer designates a group,
and a branch is created for each collection of rows in a group.
See the tar_group() function to see how you can
create the special tar_group column with dplyr::group_by().
Character of length 1, what to do if the target
runs into an error. If "stop", the whole pipeline stops
and throws an error. If "continue", the error is recorded,
but the pipeline keeps going.
Character of length 1, memory strategy.
If "persistent", the target stays in memory
until the end of the pipeline.
If "transient", the target gets unloaded
after every new target completes.
Either way, the target gets automatically loaded into memory
whenever another target needs the value.
Character of length 1, only relevant to
tar_make_clustermq() and tar_make_future(). If "remote",
the target builds on a remote parallel worker. If "local",
the target builds on the host machine / process managing the pipeline.
Numeric of length 1 between 0 and 1. Controls which targets get deployed first when multiple competing targets are ready simultaneously. Targets with priorities closer to 1 get built earlier.
A named list of computing resources. Uses:
Template file wildcards for future::future() in tar_make_future().
Template file wildcards clustermq::workers() in tar_make_clustermq().
Custom curl handle if format = "url",
e.g. resources = list(handle = curl::new_handle()).
Custom preset for qs::qsave() if format = "qs", e.g.
resources = list(handle = "archive").
Custom compression level for fst::write_fst() if
format is "fst", "fst_dt", or "fst_tbl", e.g.
resources = list(compress = 100).
AWS bucket and prefix for the "aws_" formats, e.g.
resources = list(bucket = "your-bucket", prefix = "folder/name").
bucket is required for AWS formats. See the cloud computing chapter
of the manual for details.
Character of length 1, only relevant to
tar_make_clustermq() and tar_make_future().
If "local", the target's return value is sent back to the
host machine and saved locally. If "remote", the remote worker
saves the value.
Character of length 1, only relevant to
tar_make_clustermq() and tar_make_future().
If "local", the target's dependencies are loaded on the host machine
and sent to the remote worker before the target builds.
If "remote", the remote worker loads the targets dependencies.
An optional object from tar_cue() to customize the
rules that decide whether the target is up to date.
A target object. Users should not modify these directly,
just feed them to tar_pipeline() in your _targets.R file.
# NOT RUN {
# The following are equivalent.
y <- tar_target(y, sqrt(x), pattern = map(x))
y <- tar_target_raw("y", expression(sqrt(x)), expression(map(x)))
# Programmatically create a chain of interdependent targets
target_list <- lapply(seq_len(4), function(i) {
tar_target_raw(
letters[i + 1],
substitute(do_something(x), env = list(x = rlang::sym(letters[i])))
)
})
print(target_list[[1]])
print(target_list[[2]])
# }
Run the code above in your browser using DataLab