Framework for the reproducible processing of neuroimaging data with deep learning methods
interpret —-save_nifti
issue,--fsdp
option--valid_longitudinal
option to allow the validation on longitudinal data,--fully_sharded_data_parallelism
option--emisions_calculator
option with codecarbonadapt
commandgenerate artifacts
to generate noise/contract/motion--amp
for automatic mixed precision--track_exp
option to track your parameters during training with MLflow or WandB--diagnoses
and --merged_tsv
option bug in clinicadl get-labels
get_tsv_paths
function bug.clinicadl predict
--n_proc
option in clinicadl generate
pipelines for parallelization.--split
to clinicadl predict
--size_reduction
and --size_reduction_factor
options to clinicadl train
, clinicadl predict
and clinicadl interpret
.--save_latent_space
option to clinicadl train
and clinicadl predict
.clinicadl generate trvial_motion
--extract_json
option bug in clinicadl prepare-data
.clinicadl tsvtools get-labels
error finding clinica iotools missing-modalities
output.clinicadl tsvtools get-labels
output directory.--caps_directory
option in clinicadl tsvtools get-labels
.quality-check pet-linear
.generate hypometabolic
.Resnet3D
and SqueezeExcitationCNN
.flair-linear
modality for prepare-data
command.--save_nifti
option for interpret
command.--output_dir
argument for tsvtools get-labels
command** Core: **
--acq_label
option for --tracer
.** Core: **
extract
command to prepare-data
.get-labels
, split
and kfold
commands to one TSV per split instead of one per label.tsvtool
command to tsvtools
.tsvtools getlabels
command to tsvtools get-labels
and remove the progression column in the TSV output.tsvtools get-progression
, tsvtools get-metadata
, tsvtools prepare-experiment
(split + kfold), and tsvtools adapt
.quality-check t1-linear
** Core: **
--save_tensor
and --save_nifty
options into the predict
sub-command (before save_tensor
was also subcommand`).--model_layers
to the clinicadl train list_models
subcommad
to visualize the model layers.clinicadl train list_models
command.Core:
extract
task.resume
and quality-check
functionalities.Command line:
clinicadl train
. Task is now a sub-command, and not an argument.save-tensor
to save the reconstruction outputs as nifti files.clinicadl train from_json
to train again the same experiment from a JSON file generated by ClinicaDL.--verbose
option during the training tasks. This behavior is fixed.resume
was changed.