python-microscopy 25.6.5__cp39-cp39-win_amd64.whl

PYME/recipes/output.py

@@ -0,0 +1,667 @@
+from .base import register_module, ModuleBase, OutputModule
+from .traits import Input, Output, Float, Enum, CStr, Bool, Int, Dict, Str
+
+import numpy as np
+#import pandas as pd
+import os
+from PYME.IO import tabular
+
+import logging
+logger = logging.getLogger(__name__)
+
+def _ensure_output_directory(filename):
+    dirname = os.path.dirname(filename)
+    if not os.path.exists(dirname):
+        try:
+            os.makedirs(dirname)
+        except OSError as e:
+            import errno
+            # makedirs is susceptible to race conditions - allow simultaneous
+            # creation from a separate thread / process to fail gracefully.
+            # NOTE: We check the errno rather than catching FileExistsError to maintain py2 compatibility.
+            if not e.errno == errno.EEXIST:
+                raise
+            
+
+@register_module('CSVOutput')
+class CSVOutput(OutputModule):
+    """
+    Save tabular data as csv.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the table to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+
+    When using `File` or `pyme-cluster://` schemes, we convert the data to a pandas `DataFrame` and uses the `to_csv`
+    method to save. When using `pyme-cluster:// - aggregate` we convert to a recarray and
+    use `PYME.IO.clusterResults.fileResults`.
+
+    **pyme-cluster awareness**
+    
+    csv output is cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the
+    processing node is also running a dataserver [TODO - make this more robust] .
+
+    `pyme-cluster:// - aggregate` will **append** to a file on the cluster (creating if necessary). The intended usage
+    of the aggregate scheme is allow the concatenation of summary results when a recipe is applied to multiple different
+    source files. If subsequent identification of the source file, or e.g. metadata on sample conditions, this information
+    should be added as a column to the tabular data before savine. See also `recipes.measurement.AddMetadataToMeasurements`.
+    """
+
+    inputName = Input('output')
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        import pandas as pd
+
+        out_filename = self.filePattern.format(**context)
+        v = namespace[self.inputName]
+
+        if self.scheme == 'pyme-cluster:// - aggregate':
+            from PYME.IO import clusterResults
+            clusterResults.fileResults('pyme-cluster://_aggregate_csv/' + out_filename.lstrip('/'), v.toDataFrame())
+        else:
+            out_filename = self._schemafy_filename(out_filename)
+            _ensure_output_directory(out_filename)
+            
+            if not isinstance(v, pd.DataFrame):
+                v = v.toDataFrame()
+                
+            v.to_csv(out_filename)
+
+@register_module('XLSXOutput')
+class XLSOutput(OutputModule):
+    """
+    Save tabular data as xlsx.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the table to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+
+    This casts the data to a pandas `DataFrame` and uses the `to_excel` method to save
+
+    **pyme-cluster awareness**
+    
+    xlsx output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust] .
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    filePattern = '{output_dir}/{file_stub}.xlsx'
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+        
+        v = namespace[self.inputName]
+        v.toDataFrame.to_excel(out_filename)
+
+@register_module('ImageOutput')
+class ImageOutput(OutputModule):
+    """
+    Save an image with a file type determined by extension.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the image to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+
+    This is a very thin wrapper which simply calls the `.save()` method on the `PYME.IO.image.ImageStack` object.
+
+    **pyme-cluster awareness**
+    
+    Image output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    filePattern = '{output_dir}/{file_stub}.tif'
+    extraMetadata = Dict(Str)
+    
+    def generate(self, namespace, recipe_context={}):
+        im =  namespace[self.inputName]
+        im.mdh.record_pyme_version()
+
+        for k, v in self.extraMetadata.items():
+            im.mdh[k] = v
+
+        return im
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+        
+        v = self.generate(namespace)
+        v.Save(out_filename)
+
+
+@register_module('RGBImageOutput')
+class RGBImageOutput(OutputModule):
+    """
+    Save an RGB representation of an image
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the image to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+
+    **pyme-cluster awareness**
+    
+    Image output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    filePattern = '{output_dir}/{file_stub}.png'
+    scaling = Enum(['min-max', 'percentile'])
+    scaling_factor = Float(0.95)
+    zoom = Int(1)
+    colorblindFriendly = Bool(False)
+    
+    def generate(self, namespace, recipe_context={}):
+        from PYME.IO.rgb_image import image_to_rgb, image_to_cmy
+        im = namespace[self.inputName]
+        
+        if self.colorblindFriendly:
+            return image_to_cmy(im, zoom=self.zoom, scaling=self.scaling, scaling_factor=self.scaling_factor)
+        else:
+            return image_to_rgb(im, zoom=self.zoom, scaling=self.scaling, scaling_factor=self.scaling_factor)
+            
+                
+    
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        from PIL import Image
+        
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+        
+        v = self.generate(namespace)
+        
+        Image.fromarray(v, mode='RGB').save(out_filename)
+
+@register_module('HDFOutput')
+class HDFOutput(OutputModule):
+    """
+    Save tabular data as a table in HDF5.
+
+    Parameters
+    ----------
+
+    inputVariables : dict
+        a dictionary mapping parameter names to table names
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+    
+    **pyme-cluster awareness**
+    
+    csv output is cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the
+    processing node is also running a dataserver [TODO - make this more robust] .
+
+    `pyme-cluster:// - aggregate` will **append** to a file on the cluster (creating if necessary). The intended usage
+    of the aggregate scheme is allow the concatenation of summary results when a recipe is applied to multiple different
+    source files. If subsequent identification of the source file, or e.g. metadata on sample conditions, this information
+    should be added as a column to the tabular data before savine. See also `recipes.measurement.AddMetadataToMeasurements`.
+    """
+    inputVariables = Dict(str, str)
+    filePattern = '{output_dir}/{file_stub}.hdf'
+    extraMetadata = Dict(Str)
+
+    @property
+    def inputs(self):
+        return set(self.inputVariables.keys())
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to HDF5
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        from PYME.IO.MetaDataHandler import DictMDHandler
+        out_filename = self.filePattern.format(**context)
+
+        if self.scheme == 'pyme-cluster:// - aggregate':
+            from PYME.IO import clusterResults
+
+            for name, h5_name in self.inputVariables.items():
+                v = namespace[name]
+                URI = '/'.join(['pyme-cluster:///__aggregate_h5r', out_filename.lstrip('/'), h5_name])
+                clusterResults.fileResults(URI, v.to_recarray())
+                #NOTE - aggregation does not support metadata
+        else:
+            out_filename = self._schemafy_filename(out_filename)
+            _ensure_output_directory(out_filename)
+            
+            for name, h5_name in self.inputVariables.items():
+                v = namespace[name]
+
+                mdh = DictMDHandler(getattr(v, 'mdh', None))
+                for k, val in self.extraMetadata.items():
+                    mdh[k] = val
+
+                v.to_hdf(out_filename, tablename=h5_name, metadata=mdh)
+
+    @property
+    def default_view(self):
+        import wx
+        if wx.GetApp() is None:
+            return None
+        
+        from traitsui.api import View, Item
+        from PYME.ui.custom_traits_editors import DictChoiceStrEditor
+
+        inputs, outputs, params = self.get_params()
+        
+        return View([Item(tn, editor=DictChoiceStrEditor(choices=self._namespace_keys)) for tn in inputs] +
+                    [Item('_'),] +
+                    self._view_items(params), buttons=['OK'])
+
+
+
+@register_module('ReportOutput')
+class ReportOutput(OutputModule):
+    """
+    Save an html report.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the image to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    template : basestring
+        The path / uri to the html template to use.
+
+    Notes
+    -----
+
+    This is a very thin wrapper which simply calls the `.save()` method on the `PYME.IO.image.ImageStack` object.
+
+    **pyme-cluster awareness**
+
+    report output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    filePattern = '{output_dir}/{file_stub}.html'
+
+    template = CStr('templates/report.html')
+    
+    def generate(self, namespace, recipe_context={}):
+        from PYME import reports
+        template = reports.env.get_template(self.template)
+        v = namespace[self.inputName]
+        return template.render(data=v, namespace=namespace, recipe_context=recipe_context)
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        import codecs
+
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+
+        with open(out_filename, 'wb') as f:
+            f.write(self.generate(namespace, recipe_context=context).encode('utf-8'))
+
+
+@register_module('ReportForEachOutput')
+class ReportForEachOutput(OutputModule):
+    """
+    Save an html report for each item in a sequence.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the image to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    template : basestring
+        The path / uri to the html template to use.
+
+    Notes
+    -----
+
+    This is a very thin wrapper which simply calls the `.save()` method on the `PYME.IO.image.ImageStack` object.
+
+    **pyme-cluster awareness**
+
+    report output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    inputImage = Input('')
+    filePattern = '{output_dir}/{file_stub}_{num}.html'
+    template = CStr('templates/report.html')
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        from PYME import reports
+        template = reports.env.get_template(self.template)
+
+        v = namespace[self.inputName]
+        img = namespace[self.inputImage]
+
+        for i, c in enumerate(v):
+            out_filename = self._schemafy_filename(self.filePattern.format(num=i, **context))
+            _ensure_output_directory(out_filename)
+            
+            with open(out_filename, 'w') as f:
+                f.write(template.render(data=c, img=img))
+
+
+@register_module('PlotOutput')
+class PlotOutput(OutputModule):
+    """
+    Save a matplotlib plot as encapsulated by a graphing.Plot object
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the image to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+        
+    format: enum
+        the format to save as.
+
+    Notes
+    -----
+    
+    **pyme-cluster awareness**
+
+    Image output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the processing
+    node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('output')
+    filePattern = '{output_dir}/{file_stub}.pdf'
+    format = Enum('pdf', 'png', 'svg')
+    
+    def generate(self, namespace, recipe_context={}):
+        return namespace[self.inputName]
+    
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to CSV
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+        
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+        
+        v = self.generate(namespace)
+        
+        if self.format == 'pdf':
+            v.as_pdf(out_filename)
+        elif self.format == 'png':
+            v.as_png(out_filename)
+        elif self.format == 'svg':
+            v.as_svg(out_filename)
+
+@register_module('STLOutput')
+class STLOutput(OutputModule):
+    """
+    Save a surface as an STL file.
+
+    Parameters
+    ----------
+
+    inputName : basestring
+        the name (in the recipe namespace) of the surface to save.
+
+    filePattern : basestring
+        a pattern through which the output filename is generated by variable substitution (using `str.format`)
+
+    scheme : enum
+        The storage method, one of 'File', 'pyme-cluster://' or 'pyme-cluster:// - aggregate`. File is the default
+        and saves to a file on disk.
+
+    Notes
+    -----
+
+    This is a very thin wrapper which simply calls the `.to_stl()` method on the 
+    `PYME.experimental._triangle_mesh.TriangleMesh` object.
+
+    **pyme-cluster awareness**
+    
+    STL output is semi cluster-aware. Selecting a scheme of `pyme-cluster://` will save within the cluster root on the
+    current machine, so that the results are accessible through the cluster. **NOTE:** this will **ONLY** work if the 
+    processing node is also running a dataserver [TODO - make this more robust].
+
+    `pyme-cluster:// - aggregate` is not supported.
+    """
+    inputName = Input('surf0')
+    filePattern = '{output_dir}/{file_stub}.stl'
+
+    def generate(self, namespace, recipe_context={}):
+        return namespace[self.inputName]
+
+    def save(self, namespace, context={}):
+        """
+        Save recipes output(s) to STL
+
+        Parameters
+        ----------
+        namespace : dict
+            The recipe namespace
+        context : dict
+            Information about the source file to allow pattern substitution to generate the output name. At least
+            'basedir' (which is the fully resolved directory name in which the input file resides) and
+            'file_stub' (which is the filename without any extension) should be resolved.
+
+        Returns
+        -------
+
+        """
+
+        out_filename = self._schemafy_filename(self.filePattern.format(**context))
+        _ensure_output_directory(out_filename)
+        
+        v = self.generate(namespace)
+        v.to_stl(out_filename)