view ms2deepscore_similarity.xml @ 0:58c95171a0d3 draft default tip

planemo upload for repository https://github.com/RECETOX/galaxytools/tree/master/tools/ms2deepscore commit 4bd610e0cbbcbed51a6bfb880179777fc8034fd6

<tool id="ms2deepscore_similarity" name="MS2DeepScore Similarity" version="@TOOL_VERSION@+galaxy0">
    <description>Compute similarity scores using a pre-trained MS2DeepScore model</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="creator"/>
    <expand macro="edam" />

    <requirements>
        <requirement type="package" version="@TOOL_VERSION@">ms2deepscore</requirement>
        <requirement type="package" version="@ONNX_VERSION@">onnx</requirement>
    </requirements>
    
    <command detect_errors="exit_code"><![CDATA[
        python3 ${python_wrapper}
    ]]></command>
<configfiles>
<configfile name="python_wrapper">
@init_logger@

import onnx
import torch
from ms2deepscore import MS2DeepScore
from matchms import calculate_scores
from matchms.importing import load_from_msp, scores_from_json
from ms2deepscore.models.SiameseSpectralModel import SiameseSpectralModel
from ms2deepscore.SettingsMS2Deepscore import SettingsMS2Deepscore

onnx_model = onnx.load("$model")

# Extract the initializers (weights and biases)
initializers = {init.name: onnx.numpy_helper.to_array(init) for init in onnx_model.graph.initializer}

# Convert NumPy arrays to PyTorch tensors
state_dict = {name: torch.tensor(np_array) for name, np_array in initializers.items()}

@json_load@

model = SiameseSpectralModel(settings=SettingsMS2Deepscore(**model_params))
model.load_state_dict(state_dict)
model.eval()

#if $scores.use_scores == "True"
scores_in = scores_from_json("${scores_in}")
references=scores_in.references
queries=scores_in.queries
#else
references=list(load_from_msp("$references"))
queries=list(load_from_msp("$queries"))
#end if

similarity = MS2DeepScore(model)
scores = calculate_scores(references, queries, similarity)

#if $scores.use_scores == "True"
scores_in._scores.add_dense_matrix(scores.to_array(), "MS2DeepScore")
scores_in.to_json("$similarity_scores")
#else
scores.to_json("$similarity_scores")
#end if
</configfile>
</configfiles>
    
    <inputs>
        <expand macro="input_param" />
    </inputs>
    
    <outputs>
        <data label="ms2deepscore scores of ${on_string}" name="similarity_scores" format="json"/>
    </outputs>

    <tests>
        <test expect_num_outputs="1">
            <param name="use_scores" value="False"/>
            <param name="references" value="inp_filtered_library.msp" ftype="msp"/>
            <param name="queries" value="inp_filtered_spectra.msp" ftype="msp"/>
            <param name="model" value="Trained_model.onnx" ftype="onnx"/>
            <param name="model_param" value="Model_Parameter_JSON.json" ftype="json"/>
            <output name="similarity_scores" value="msp_json_score_out.json" ftype="json" compare="sim_size" />
        </test>
        <test expect_num_outputs="1">
            <param name="use_scores" value="True"/>
            <param name="scores_in" value="ri_match_60.json" ftype="json"/>
            <param name="model" value="Trained_model.onnx" ftype="onnx"/>
            <param name="model_param" value="Model_Parameter_JSON.json" ftype="json"/>
            <output name="similarity_scores" value="usescore_json_score_out.json" ftype="json" compare="sim_size" />
        </test>
    </tests>
    
    <help>
<![CDATA[
Info
====
Use a MS2DeepScore model for spectral similarity calculation.
For security reasons, only ONNX models are supported to be used.
If you want to use your own model, you can convert it to ONNX using the code contained in the 'MS2DeepScore Model Training' tool.

About
=====
@HELP@
]]>
    </help>   
    <expand macro="citations"/>
</tool>