Command Line Interface Reference

themeda

themeda [OPTIONS] COMMAND [ARGS]...

Options

-v, --version

Prints the current version.

--install-completion <install_completion>

Install completion for the specified shell.

Options

bash | zsh | fish | powershell | pwsh

--show-completion <show_completion>

Show completion for the specified shell, to copy it or customize the installation.

Options

bash | zsh | fish | powershell | pwsh

bibliography

themeda bibliography [OPTIONS]

bibtex

themeda bibtex [OPTIONS]

infer

themeda infer [OPTIONS]

Options

--gpu, --no-gpu

Whether or not to use a GPU for processing if available.

Default

True

--pretrained <pretrained>

The location (URL or filepath) of a pretrained model.

--reload, --no-reload

Should the pretrained model be downloaded again if it is online and already present locally.

Default

False

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--probabilities <probabilities>
--argmax <argmax>

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir

lr-finder

themeda lr-finder [OPTIONS]

Options

--plot-filename <plot_filename>
--start-lr <start_lr>
Default

1e-07

--end-lr <end_lr>
Default

10

--iterations <iterations>
Default

100

--fp16, --no-fp16

Whether or not the floating-point precision of learner should be set to 16 bit.

Default

True

--output-dir <output_dir>

The location of the output directory.

Default

./outputs

--weight-decay <weight_decay>

The amount of weight decay. If None then it uses the default amount of weight decay in fastai.

--metric-dir <metric_dir>

The location to save data for the metrics.

--persistence, --no-persistence
Default

False

--input <input>

Required The input data types.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--output <output>

The output data types. If not given, then the outputs are the same as the inputs.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--validation-subset <validation_subset>
Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--label-smoothing <label_smoothing>
Default

0.0

--emd-loss, --no-emd-loss
Default

False

--hierarchical-embedding, --no-hierarchical-embedding
Default

False

--embedding-size <embedding_size>

The number of embedding dimensions.

Default

16

--cnn-kernel <cnn_kernel>
Default

15

--cnn-size <cnn_size>
Default

64

--cnn-layers <cnn_layers>
Default

1

--layers <layers>
Default

4

--growth-factor <growth_factor>
Default

2.0

--padding-mode <padding_mode>
Default

reflect

--temporal-processor-type <temporal_processor_type>
Default

LSTM

Options

LSTM | GRU | TRANSFORMER | NONE

--temporal-layers <temporal_layers>
Default

2

--temporal-size <temporal_size>
Default

32

--transformer-heads <transformer_heads>
Default

8

--transformer-layers <transformer_layers>
Default

4

--transformer-positional-encoding, --no-transformer-positional-encoding
Default

True

--unet, --no-unet
Default

False

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir

show-batch

themeda show-batch [OPTIONS]

Options

--output-path <output_path>

A location to save the HTML which summarizes the batch.

--input <input>

Required The input data types.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--output <output>

The output data types. If not given, then the outputs are the same as the inputs.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--validation-subset <validation_subset>
Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--label-smoothing <label_smoothing>
Default

0.0

--emd-loss, --no-emd-loss
Default

False

--hierarchical-embedding, --no-hierarchical-embedding
Default

False

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir

train

themeda train [OPTIONS]

Options

--distributed, --no-distributed

If the learner is distributed.

Default

False

--fp16, --no-fp16

Whether or not the floating-point precision of learner should be set to 16 bit.

Default

True

--output-dir <output_dir>

The location of the output directory.

Default

./outputs

--weight-decay <weight_decay>

The amount of weight decay. If None then it uses the default amount of weight decay in fastai.

--metric-dir <metric_dir>

The location to save data for the metrics.

--persistence, --no-persistence
Default

False

--input <input>

Required The input data types.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--output <output>

The output data types. If not given, then the outputs are the same as the inputs.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--validation-subset <validation_subset>
Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--label-smoothing <label_smoothing>
Default

0.0

--emd-loss, --no-emd-loss
Default

False

--hierarchical-embedding, --no-hierarchical-embedding
Default

False

--embedding-size <embedding_size>

The number of embedding dimensions.

Default

16

--cnn-kernel <cnn_kernel>
Default

15

--cnn-size <cnn_size>
Default

64

--cnn-layers <cnn_layers>
Default

1

--layers <layers>
Default

4

--growth-factor <growth_factor>
Default

2.0

--padding-mode <padding_mode>
Default

reflect

--temporal-processor-type <temporal_processor_type>
Default

LSTM

Options

LSTM | GRU | TRANSFORMER | NONE

--temporal-layers <temporal_layers>
Default

2

--temporal-size <temporal_size>
Default

32

--transformer-heads <transformer_heads>
Default

8

--transformer-layers <transformer_layers>
Default

4

--transformer-positional-encoding, --no-transformer-positional-encoding
Default

True

--unet, --no-unet
Default

False

--epochs <epochs>

The number of epochs.

Default

20

--freeze-epochs <freeze_epochs>

The number of epochs to train when the learner is frozen and the last layer is trained by itself. Only if fine_tune is set on the app.

Default

3

--learning-rate <learning_rate>

The base learning rate (when fine tuning) or the max learning rate otherwise.

Default

0.0001

--project-name <project_name>

The name for this project for logging purposes.

--run-name <run_name>

The name for this particular run for logging purposes.

--run-id <run_id>

A unique ID for this particular run for logging purposes.

--notes <notes>

A longer description of the run for logging purposes.

--tag <tag>

A tag for logging purposes. Multiple tags can be added each introduced with –tag.

--wandb, --no-wandb

Whether or not to use ‘Weights and Biases’ for logging.

Default

False

--wandb-mode <wandb_mode>

The mode for ‘Weights and Biases’.

Default

online

--wandb-dir <wandb_dir>

The location for ‘Weights and Biases’ output.

--wandb-entity <wandb_entity>

An entity is a username or team name where you’re sending runs.

--wandb-group <wandb_group>

Specify a group to organize individual runs into a larger experiment.

--wandb-job-type <wandb_job_type>

Specify the type of run, which is useful when you’re grouping runs together into larger experiments using group.

--mlflow, --no-mlflow

Whether or not to use MLflow for logging.

Default

False

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir

tune

themeda tune [OPTIONS]

Options

--runs <runs>

The number of runs to attempt to train the model.

Default

1

--engine <engine>

The optimizer to use to perform the hyperparameter tuning. Options: wandb, optuna, skopt.

Default

skopt

--id <id>

The ID of this hyperparameter tuning job. If using wandb, then this is the sweep id. If using optuna, then this is the storage. If using skopt, then this is the file to store the results.

Default

--name <name>

An informative name for this hyperparameter tuning job. If empty, then it creates a name from the project name.

Default

--method <method>

The sampling method to use to perform the hyperparameter tuning. By default it chooses the default method of the engine.

Default

--min-iter <min_iter>

The minimum number of iterations if using early termination. If left empty, then early termination is not used.

--seed <seed>

A seed for the random number generator.

--distributed, --no-distributed

If the learner is distributed.

Default

False

--fp16, --no-fp16

Whether or not the floating-point precision of learner should be set to 16 bit.

Default

True

--output-dir <output_dir>

The location of the output directory.

Default

./outputs

--weight-decay <weight_decay>

The amount of weight decay. If None then it uses the default amount of weight decay in fastai.

--metric-dir <metric_dir>

The location to save data for the metrics.

--persistence, --no-persistence
Default

False

--input <input>

Required The input data types.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--output <output>

The output data types. If not given, then the outputs are the same as the inputs.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--validation-subset <validation_subset>
Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--label-smoothing <label_smoothing>
Default

0.0

--emd-loss, --no-emd-loss
Default

False

--hierarchical-embedding, --no-hierarchical-embedding
Default

False

--embedding-size <embedding_size>

The number of embedding dimensions.

Default

16

--cnn-kernel <cnn_kernel>
Default

15

--cnn-size <cnn_size>
Default

64

--cnn-layers <cnn_layers>
Default

1

--layers <layers>
Default

4

--growth-factor <growth_factor>
Default

2.0

--padding-mode <padding_mode>
Default

reflect

--temporal-processor-type <temporal_processor_type>
Default

LSTM

Options

LSTM | GRU | TRANSFORMER | NONE

--temporal-layers <temporal_layers>
Default

2

--temporal-size <temporal_size>
Default

32

--transformer-heads <transformer_heads>
Default

8

--transformer-layers <transformer_layers>
Default

4

--transformer-positional-encoding, --no-transformer-positional-encoding
Default

True

--unet, --no-unet
Default

False

--epochs <epochs>

The number of epochs.

Default

20

--freeze-epochs <freeze_epochs>

The number of epochs to train when the learner is frozen and the last layer is trained by itself. Only if fine_tune is set on the app.

Default

3

--learning-rate <learning_rate>

The base learning rate (when fine tuning) or the max learning rate otherwise.

Default

0.0001

--project-name <project_name>

The name for this project for logging purposes.

--run-name <run_name>

The name for this particular run for logging purposes.

--run-id <run_id>

A unique ID for this particular run for logging purposes.

--notes <notes>

A longer description of the run for logging purposes.

--tag <tag>

A tag for logging purposes. Multiple tags can be added each introduced with –tag.

--wandb, --no-wandb

Whether or not to use ‘Weights and Biases’ for logging.

Default

False

--wandb-mode <wandb_mode>

The mode for ‘Weights and Biases’.

Default

online

--wandb-dir <wandb_dir>

The location for ‘Weights and Biases’ output.

--wandb-entity <wandb_entity>

An entity is a username or team name where you’re sending runs.

--wandb-group <wandb_group>

Specify a group to organize individual runs into a larger experiment.

--wandb-job-type <wandb_job_type>

Specify the type of run, which is useful when you’re grouping runs together into larger experiments using group.

--mlflow, --no-mlflow

Whether or not to use MLflow for logging.

Default

False

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir

validate

themeda validate [OPTIONS]

Options

--gpu, --no-gpu

Whether or not to use a GPU for processing if available.

Default

True

--persistence, --no-persistence
Default

False

--metric-dir <metric_dir>
--pretrained <pretrained>

The location (URL or filepath) of a pretrained model.

--reload, --no-reload

Should the pretrained model be downloaded again if it is online and already present locally.

Default

False

--input <input>

Required The input data types.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--output <output>

The output data types. If not given, then the outputs are the same as the inputs.

Options

land_cover | tmax | rain | elevation | land_use | fire_scar_early | fire_scar_late | soil_ece | soil_depth | soil_clay

--roi <roi>

The Region of Interest.

Default

savanna

Options

savanna | australia

--base-dir <base_dir>

Required The base directory for the preprocessed data.

--start-year <start_year>

The start year.

Default

1988

--end-year <end_year>

The end year (inclusive).

Default

2018

--max-chiplets <max_chiplets>
--max-years <max_years>
--batch-size <batch_size>

The batch size.

Default

1

--validation-subset <validation_subset>
Default

1

--pad-size <pad_size>
Default

0

--base-size <base_size>
Default

160

--label-smoothing <label_smoothing>
Default

0.0

--emd-loss, --no-emd-loss
Default

False

--hierarchical-embedding, --no-hierarchical-embedding
Default

False

Environment variables

THEMEDA_PREPROC_BASE_OUTPUT_DIR

Provide a default for --base-dir