Create a model to recommend tools tool metadata

Miscellaneous

Name: Create a model to recommend tools

Description: using deep learning

Id: create_tool_recommendation_model

Guid: toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.5

Version: 0.0.5

Version command string: echo "@VERSION@"

Display in tool panel: True

Version lineage of this tool (guids ordered most recent to oldest)

toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.5 (this tool)

toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.4

toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.3

toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.2

toolshed.g2.bx.psu.edu/repos/bgruening/create_tool_recommendation_model/create_tool_recommendation_model/0.0.1

create_tool_recommendation_model

Requirements (dependencies defined in the <requirements> tag set)

Requirements:

name

version

type

python

3.9.12

package

tensorflow

2.9.1

package

keras

2.9.0

package

scikit-learn

1.0.2

package

pandas

1.4.2

package

h5py

3.6.0

package

Additional information about this tool

Command:

python '$__tool_directory__/main.py'

            --workflow_file '$input_tabular_workflows'
            --tool_usage_file '$input_tabular_tool_usage'
            --cutoff_date '$data_parameters.input_cutoff_date'
            --maximum_path_length '$data_parameters.input_maximum_path_length'
            --te_share '$training_parameters.te_share'
            --tr_batch_size '$training_parameters.tr_batch_size'
            --te_batch_size '$training_parameters.te_batch_size'
            --n_train_iter '$training_parameters.n_train_iter'
            --tr_logging_step '$training_parameters.tr_logging_step'
            --te_logging_step '$training_parameters.te_logging_step'
            --n_heads '$nn_parameters.n_heads'
            --n_embed_dim '$nn_parameters.n_embed_dim'
            --n_feed_forward_dim '$nn_parameters.n_feed_forward_dim'
            --dropout '$nn_parameters.dropout'
            --learning_rate '$nn_parameters.learning_rate'
            --output_model '$outfile_model'

Interpreter: None

Forces a history refresh: False

Functional tests

name

inputs

outputs

required files

Test-1

name: value

test_workflow_connections
test_tool_usage
value

Test-2

input_tabular_workflows: test_workflow_connections
input_tabular_tool_usage: test_tool_usage
training_parameters|n_train_iter: 20
training_parameters|tr_batch_size: 4
training_parameters|te_batch_size: 4
training_parameters|tr_logging_step: 10
training_parameters|te_logging_step: 10
training_parameters|te_share: 0.4
nn_parameters|n_heads: 2
nn_parameters|n_embed_dim: 32
nn_parameters|n_feed_forward_dim: 32
nn_parameters|dropout: 0.05
nn_parameters|learning_rate: 0.0001

name: value

test_workflow_connections
test_tool_usage
value

Test-3

input_tabular_workflows: test_workflow_connections
input_tabular_tool_usage: test_tool_usage
training_parameters|te_share: 0.0,0.1