brainscore-vision 2.1__py3-none-any.whl

brainscore_vision/data/david2004/data_packaging/lib/glob/glob.m

@@ -0,0 +1,472 @@
+%% Expand wildcards for files and directory names
+%
+%   Pattern matching of file and directory names, based on wildcard
+%   characters. This function is similar to wildcard expansion performed by
+%   the Unix shell and Python glob.glob function, but it can handle more
+%   types of wildcards.
+%
+%   [LIST, ISDIR] = glob(FILESPEC)
+%       returns cell array LIST with files or directories that match the
+%       path specified by string FILESPEC. Wildcards may be used for
+%       basenames and for the directory parts. If FILESPEC contains
+%       directory parts, then these will be included in LIST.
+%       ISDIR is a boolean, the same size as LIST that is true for
+%       directories in LIST.
+%
+%       Following wildcards can be used:
+%           *        match zero or more characters
+%           ?        match any single character
+%           [ab12]   match one of the specified characters
+%           [^ab12]  match none of the specified characters
+%           [a-z]    match one character in range of characters
+%           {a,b,c}  matches any one of strings a, b or c
+%
+%           all above wildcards do not match a file separator.
+%
+%           **       match zero or more characters including file separators.
+%                    This can be used to match zero or more directory parts
+%                    and will recursively list matching names.
+%
+%       The differences between GLOB and DIR:
+%           * GLOB supports wildcards for directories.
+%           * GLOB returns the directory part of FILESPEC.
+%           * GLOB returns a cell array of matching names.
+%           * GLOB does not return hidden files and directories that start 
+%             with '.' unless explicitly specified in FILESPEC.
+%           * GLOB does not return '.' and '..' unless explicitly specified 
+%             in FILESPEC.
+%           * GLOB adds a trailing file separator to directory names.
+%           * GLOB does not return the contents of a directory when
+%             a directory is specified. To return contents of a directory,
+%             add a trailing '/*'.
+%           * GLOB returns only directory names when a trailing file
+%             separator is specified.
+%           * On Windows GLOB is not case sensitive, but it returns
+%             matching names exactely in the case as they are defined on
+%             the filesystem. Case of host and sharename of a UNC path and
+%             case of drive letters will be returned as specified in 
+%             FILESPEC.
+%
+%   glob(FILESPEC, '-ignorecase')
+%        Default GLOB is case sensitive on Unix. With option '-ignorecase'
+%        FILESPEC matching is not case sensitive. On Windows, GLOB always
+%        ignores the case. This option can be abbreviated to '-i'.
+%
+% Examples:
+%       glob *.m        list all .m files in current directory.
+%
+%       glob baz/*      list all files and directories in subdirectory 'baz'.
+%
+%       glob b*/*.m     list all .m files in subdirectory names starting
+%                       with 'b'. The list will include the names of the
+%                       matching subdirectories.
+%
+%       glob ?z*.m      list all .m files where the second character
+%                       is 'z'.
+%
+%       glob baz.[ch]   matches baz.c and baz.h
+%
+%       glob test.[^ch] matches test.a but not test.c or test.h
+%
+%       glob demo.[a-c] matches demo.a, demo.b, and demo.c
+%
+%       glob test.{foo,bar,baz} matches test.foo, test.bar, and test.baz
+%
+%       glob .*         list all hidden files in current directory,
+%                       excluding '.' and '..'
+%
+%       glob */         list all subdirectories.
+%
+%       glob **         recursively list all files and directories,
+%                       starting in current directory (current directory
+%                       name, hidden files and hidden directories are 
+%                       excluded).
+%
+%       glob **.m       list all m-files anywhere in directory tree,
+%                       including m-files in current directory. This
+%                       is equivalent with '**/*.m'.
+%
+%       glob foo/**/    recursively list all directories, starting in
+%                       directory 'foo'.
+%
+%       glob **/.svn/   list all .svn directories in directory tree.
+%
+%       glob **/.*/**   recursively list all files in hidden directories
+%                       only.
+%
+%       [r,d]=glob('**')
+%       r(~d)           get all files in directory tree.
+%
+% Known limitation:
+%       When using '**', symbolic linked directories or junctions may cause
+%       an infinite loop.
+%
+% See also dir
+
+%% Last modified
+%   $Date: 2013-02-02 18:41:41 +0100 (Sat, 02 Feb 2013) $
+%   $Author: biggelar $
+%   $Rev: 12966 $
+
+%% History
+%   2013-02-02  biggelar    submitted to Matlab Central
+%   2013-01-11  biggelar    add {} wildcards
+%   2013-01-02  biggelar    Created
+
+%% Copyright (c) 2013, Peter van den Biggelaar
+% All rights reserved.
+% 
+% Redistribution and use in source and binary forms, with or without 
+% modification, are permitted provided that the following conditions are 
+% met:
+% 
+%     * Redistributions of source code must retain the above copyright 
+%       notice, this list of conditions and the following disclaimer.
+%     * Redistributions in binary form must reproduce the above copyright 
+%       notice, this list of conditions and the following disclaimer in 
+%       the documentation and/or other materials provided with the distribution
+%       
+% THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 
+% AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 
+% IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 
+% ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE 
+% LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 
+% CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 
+% SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 
+% INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 
+% CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 
+% ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 
+% POSSIBILITY OF SUCH DAMAGE.
+
+
+% ------------------------------------------------------------------------
+function [LIST, ISDIR] = glob(FILESPEC, ignorecase)
+
+%% check FILESPEC input
+if ischar(FILESPEC)
+    if isempty(FILESPEC)
+        % return when FILESPEC is empty
+        LIST = cell(0);
+        ISDIR = false(0);
+        return
+    elseif size(FILESPEC,1)>1
+        error('glob:invalidInput', 'FILESPEC must be a single string.')
+    end
+else
+    error('glob:invalidInput', 'FILESPEC must be a string.')
+end    
+
+%% check ignorecase option
+if nargin==2
+    if ischar(ignorecase)
+        % ignore case when option is specified; must be at least 2 characters long
+        if strncmp(ignorecase, '-ignorecase', max(numel(ignorecase),2));
+            ignorecase = true;
+        else
+            error('glob:invalidOption', 'Invalid option.') 
+        end    
+    else
+        error('glob:invalidOption', 'Invalid option.')
+    end    
+else
+    % Windows is not case sensitive
+    % Unix is case sensitive
+    ignorecase = ispc;
+end
+
+%% define function handle to regular expression function for the specified case sensitivity
+if ignorecase
+    regexp_fhandle = @regexpi;
+else
+    regexp_fhandle = @regexp;
+end
+
+%% only use forward slashes as file separator to prevent escaping backslashes in regular expressions
+filespec = strrep(FILESPEC, '\', '/');
+
+%% split pathroot part from FILESPEC
+if strncmp(filespec, '//',2)
+    if ispc
+        % FILESPEC specifies a UNC path
+        % It is not allowed to get a directory listing of share names of a 
+        % host with the DIR command.
+        % pathroot will contains e.g. //host/share/
+        pathroot = regexprep(filespec, '(^//+[^/]+/[^/]+/)(.*)', '$1');
+        filespec = regexprep(filespec, '(^//+[^/]+/[^/]+/)(.*)', '$2');
+    else
+        % for Unix, multiple leading file separators are equivalent with a single file separator
+        filespec = regexprep(filespec, '^/*', '/');
+    end
+elseif strncmp(filespec, '/', 1)
+    % FILESPEC specifies a absolute path
+    pathroot = '/';
+    filespec(1) = [];
+elseif ispc && numel(filespec)>=2 && filespec(2)==':'
+    % FILESPEC specifies a absolute path starting with a drive letter
+    % check for a fileseparator after ':'. e.g. 'C:\'
+    if numel(filespec)<3 || filespec(3)~='/'
+        error('glob:invalidInput','Drive letter must be followed by '':\''.')
+    end
+    pathroot = filespec(1:3);
+    filespec(1:3) = [];
+else
+    % FILESPEC specifies a relative path
+    pathroot = './';
+end
+
+%% replace multiple file separators by a single file separator
+filespec = regexprep(filespec, '/+', '/');
+
+%% replace 'a**' with 'a*/**', where 'a' can be any character but not '/'
+filespec = regexprep(filespec, '([^/])(\.\*\.\*)', '$1\*/$2');
+%% replace '**a' with '**/*a', where a can be any character but not '/'
+filespec = regexprep(filespec, '(\.\*\.\*)([^/])', '$1/\*$2');
+
+%% split filespec into chunks at file separator
+chunks = strread(filespec, '%s', 'delimiter', '/'); %#ok<FPARK>
+
+%% add empty chunk at the end when filespec ends with a file separator
+if ~isempty(filespec) && filespec(end)=='/'
+    chunks{end+1} = '';
+end
+
+%% translate chunks to regular expressions
+for i=1:numel(chunks)
+    chunks{i} = glob2regexp(chunks{i});
+end
+
+%% determine file list using LS_REGEXP
+% this function requires that PATHROOT does not to contain any wildcards
+if ~isempty(chunks)
+    list = ls_regexp(regexp_fhandle, pathroot, chunks{1:end});
+else
+    list = {pathroot};
+end
+
+if strcmp(pathroot, './')
+    % remove relative pathroot from result
+    list = regexprep(list, '^\./', '');
+end
+
+if nargout==2
+    % determine directories by checking for '/' at the end
+    I = regexp(list', '/$');
+    ISDIR = ~cellfun('isempty', I);
+end
+
+%% convert to standard file separators for PC
+if ispc
+    list = strrep(list, '/', '\');
+end
+
+%% return output
+if nargout==0
+    if ~isempty(list)
+        % display list
+        disp(char(list))
+    else
+        disp(['''' FILESPEC ''' not found.']);
+    end    
+else
+    LIST = list';
+end
+
+
+% ------------------------------------------------------------------------
+function regexp_str = glob2regexp(glob_str)
+%% translate glob_str to regular expression string
+
+% initialize
+regexp_str  = '';
+in_curlies  = 0;        % is > 0 within curly braces
+
+% handle characters in glob_str one-by-one
+for c = glob_str
+        
+    if any(c=='.()|+^$@%')
+        % escape simple special characters
+        regexp_str = [regexp_str '\' c]; %#ok<AGROW>
+            
+    elseif c=='*'
+        % '*' should not match '/'
+        regexp_str = [regexp_str '[^/]*']; %#ok<AGROW>
+        
+    elseif c=='?'
+        % '?' should not match '/'
+        regexp_str = [regexp_str '[^/]']; %#ok<AGROW>
+        
+    elseif c=='{'
+        regexp_str = [regexp_str '(']; %#ok<AGROW>
+        in_curlies = in_curlies+1;    
+
+    elseif c=='}' && in_curlies
+        regexp_str = [regexp_str ')']; %#ok<AGROW>
+        in_curlies = in_curlies-1;    
+
+    elseif c==',' && in_curlies
+        regexp_str = [regexp_str '|']; %#ok<AGROW>
+            
+    else                    
+        regexp_str = [regexp_str c]; %#ok<AGROW>
+    end
+end
+
+% replace original '**' (that has now become '[^/]*[^/]*') with '.*.*'  
+regexp_str = strrep(regexp_str, '[^/]*[^/]*', '.*.*');
+
+
+
+% ------------------------------------------------------------------------
+function L = ls_regexp(regexp_fhandle, path, varargin)
+% List files that match PATH/r1/r2/r3/... where PATH is a string without
+% any wildcards and r1..rn are regular expresions that contain the parts of
+% a filespec between the file separators.
+% L is a cell array with matching file or directory names.
+% REGEXP_FHANDLE contain a file handle to REGEXP or REGEXPI depending
+% on specified case sensitivity.
+    
+% if first regular expressions contains '**', examine complete file tree
+if nargin>=3 && any(regexp(varargin{1}, '\.\*\.\*'))
+    L = ls_regexp_tree(regexp_fhandle, path, varargin{:});
+    
+else
+    % get contents of path
+    list = dir(path);
+    
+    if nargin>=3
+        if strcmp(varargin{1},'\.') || strcmp(varargin{1},'\.\.')
+            % keep explicitly specified '.' or '..' in first regular expression
+            if ispc && ~any(strcmp({list.name}, '.'))
+                % fix strange windows behaviour: root of a volume has no '.' and '..'
+                list(end+1).name = '.';
+                list(end).isdir = true;
+                list(end+1).name = '..';
+                list(end).isdir = true;                
+            end    
+        else
+            % remove '.' and '..'
+            list(strcmp({list.name},'.')) = [];
+            list(strcmp({list.name},'..')) = [];
+     
+            % remove files starting with '.' specified in first regular expression
+            if ~strncmp(varargin{1},'\.',2)
+                % remove files starting with '.' from list
+                list(strncmp({list.name},'.',1))  = [];
+            end    
+        end
+    end
+    
+    % define shortcuts
+    list_isdir = [list.isdir];
+    list_name = {list.name};
+    
+    L = {};  % initialize
+    if nargin==2    % no regular expressions
+        %% return filename
+        if ~isempty(list_name)
+            % add a trailing slash to directories
+            trailing_fsep = repmat({''}, size(list_name));
+            trailing_fsep(list_isdir) = {'/'};
+            L = strcat(path, list_name, trailing_fsep);
+        end
+
+    elseif nargin==3    % last regular expression
+        %% return list_name matching regular expression
+        I = regexp_fhandle(list_name, ['^' varargin{1} '$']);
+        I = ~cellfun('isempty', I);
+        list_name = list_name(I);
+        list_isdir = list_isdir(I);
+        if ~isempty(list_name)
+            % add a trailing slash to directories
+            trailing_fsep = repmat({''}, size(list_name));
+            trailing_fsep(list_isdir) = {'/'};
+            L = strcat(path, list_name, trailing_fsep);
+        end
+        
+    elseif nargin==4 && isempty(varargin{2})    
+        %% only return directories when last regexp is empty
+        % return list_name matching regular expression and that are directories
+        I = regexp_fhandle(list_name, ['^' varargin{1} '$']);
+        I = ~cellfun('isempty', I);
+        % only return directories
+        list_name = list_name(I);
+        list_isdir = list_isdir(I);
+        if any(list_isdir)
+            % add a trailing file separator
+            L = strcat(path, list_name(list_isdir), '/');
+        end            
+    else
+        %% traverse for list_name matching regular expression
+        I = regexp_fhandle(list_name, ['^' varargin{1} '$']);
+        I = ~cellfun('isempty', I);
+        for name = list_name(I)
+            L = [L   ls_regexp(regexp_fhandle, [path char(name) '/'], varargin{2:end})]; %#ok<AGROW>
+        end
+    end
+end
+
+
+% ------------------------------------------------------------------------
+function L = ls_regexp_tree(regexp_fhandle, path, varargin)
+% use this function when first argument of varargin contains '**'
+
+% build list of complete directory tree
+% if any regexp starts with '\.', keep hidden files and directories
+I = regexp(varargin, '^\\\.');
+I = ~cellfun('isempty', I);
+keep_hidden = any(I);
+list = dir_recur(path, keep_hidden);
+L = {list.name};
+
+% make one regular expression of all individual regexps
+expression = [regexptranslate('escape',path) sprintf('%s/', varargin{1:end-1}) varargin{end}];
+
+% note that /**/ must also match zero directories
+% replace '/**/' with (/**/|/)
+expression = regexprep(expression, '/\.\*\.\*/', '(/\.\*\.\*/|/)');
+
+% return matching names
+if ~isempty(varargin{end})
+    % determing matching names ignoring trailing '/'
+    L_no_trailing_fsep = regexprep(L, '/$', '');
+    I = regexp_fhandle(L_no_trailing_fsep, ['^' expression '$']);
+else
+    % determing matching names including trailing '/'
+    I = regexp_fhandle(L, ['^' expression '$']);
+end
+I = cellfun('isempty', I);
+L(I) = [];
+
+
+
+% ------------------------------------------------------------------------
+function d = dir_recur(startdir,keep_hidden)
+%% determine recursive directory contents
+
+% get directory contents
+d = dir(startdir);
+
+% remove hidden files
+if keep_hidden
+    % only remove '.' and '..'
+    d(strcmp({d.name},'.'))  = [];
+    d(strcmp({d.name},'..')) = [];
+else
+    % remove all hidden files and directories
+    d(strncmp({d.name},'.',1)) = [];
+end
+
+if ~isempty(d)
+    % add trailing fileseparator to directories
+    trailing_fsep = repmat({''}, size(d));
+    trailing_fsep([d.isdir]) = {'/'};
+    
+    % prefix startdir to name and postfix fileseparator for directories
+    dname = strcat(startdir, {d.name}, trailing_fsep');
+    [d(:).name] = deal(dname{:});
+    
+    % recurse into subdirectories
+    for subd = {d([d.isdir]).name}
+        d = [d; dir_recur(char(subd), keep_hidden)]; %#ok<AGROW>
+    end
+end

brainscore_vision/data/david2004/data_packaging/lib/glob/license.txt

@@ -0,0 +1,27 @@
+Copyright (c) 2013, peter van den Biggelaar
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in
+      the documentation and/or other materials provided with the distribution
+    * Neither the name of the ASML nor the names
+      of its contributors may be used to endorse or promote products derived
+      from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.

brainscore_vision/data/david2004/data_packaging/xr_align_debug.py

@@ -0,0 +1,137 @@
+import glob
+import os
+import re
+import string
+
+import numpy as np
+import pandas as pd
+import xarray as xr
+import mkgu
+import mkgu.assemblies
+from mkgu.knownfile import KnownFile as kf
+
+
+def align_debug():
+    v2_base_path = "/braintree/data2/active/users/jjpr/mkgu_packaging/crcns/v2-1"
+    nc_files = sorted(glob.glob(os.path.join(v2_base_path, "*/*/*.nc"), recursive=True))
+    gd_arrays = []
+    nonzeros_raw = []
+    for f in (nc_files[0], nc_files[5]):
+        print(f)
+        gd_array = xr.open_dataarray(f)
+        # gd_array = gd_array.T.rename({"image_file_name": "presentation"})
+        # gd_array.coords["presentation_id"] = ("presentation", range(gd_array.shape[1]))
+        # gd_array = gd_array.rename({"image_file_name": "presentation"})
+        # gd_array.coords["presentation_id"] = ("presentation", range(gd_array.shape[0]))
+        gd_array.coords["presentation_id"] = ("image_file_name", range(gd_array.shape[0]))
+        # gd_array.coords["neuroid_id"] = ("neuroid", gd_array["neuroid"].values)
+        # df_massage = pd.DataFrame(list(map(massage_file_name, gd_array["presentation"].values)))
+        # for column in df_massage.columns:
+        #     gd_array.coords[column] = ("presentation", df_massage[column])
+        # gd_array.reset_index(["neuroid", "presentation"], drop=True, inplace=True)
+        gd_array.reset_index("category_name", drop=True, inplace=True)
+        mkgu.assemblies.gather_indexes(gd_array)
+        gd_arrays.append(gd_array)
+        nonzeros_raw.append(np.nonzero(~np.isnan(gd_array)))
+    print("nonzeros_raw: ")
+    print(nonzeros_raw)
+    align_test = xr.align(*gd_arrays, join="outer")
+    nonzeros_aligned = [np.nonzero(~np.isnan(da)) for da in align_test]
+    print("nonzeros_aligned: ")
+    print(nonzeros_aligned)
+    assert nonzeros_raw[0].shape == nonzeros_aligned[0].shape
+
+
+def massage_file_name(file_name):
+    split = re.split("\\\\|/", file_name)
+    split = [t for t in split if t]
+    relative_path = os.path.join(*split[-5:])
+    full_path = os.path.join("/", *split)
+    sha1 = kf(full_path).sha1
+    result = {
+        "image_file_path_original": relative_path,
+        "image_id": sha1
+    }
+    return result
+
+
+def align_bug_reproduce():
+    dims = ("x", "y")
+    shape = (10, 5)
+    das = []
+    for j in (0, 1):
+        data = np.full(shape, np.nan, dtype="float64")
+        for i in range(shape[0]):
+            data[i, i % shape[1]] = float(i)
+        coords_d = {
+            "ints": ("x", range(j*shape[0], (j+1)*shape[0])),
+            "nans": ("x", np.array([np.nan] * shape[0], dtype="float64")),
+            "lower": ("y", list(string.ascii_lowercase[:shape[1]]))
+        }
+        da = xr.DataArray(data=data, dims=dims, coords=coords_d)
+        da.set_index(append=True, inplace=True, x=["ints", "nans"], y=["lower"])
+        das.append(da)
+    nonzeros_raw = [np.nonzero(~np.isnan(da)) for da in das]
+    print("nonzeros_raw: ")
+    print(nonzeros_raw)
+    aligned = xr.align(*das, join="outer")
+    nonzeros_aligned = [np.nonzero(~np.isnan(da)) for da in aligned]
+    print("nonzeros_aligned: ")
+    print(nonzeros_aligned)
+    assert nonzeros_raw[0].shape == nonzeros_aligned[0].shape
+
+
+def align_bug_reproduce_old():
+    dims = ("x", "y")
+    coords_d = {"x": ("tens", "negative", "nans"), "y": ("lower", "upper")}
+
+    shape6 = (15, 10)
+    data6 = np.full(shape6, np.nan, dtype="float64")
+    for i in range(shape6[0]):
+        data6[i, i % shape6[1]] = float(i)
+    coords6 = {
+        "lower": ("y", list(string.ascii_lowercase[:shape6[1]])),
+        "upper": ("y", [c + string.ascii_uppercase for c in list(string.ascii_uppercase[:shape6[1]])]),
+        "tens": ("x", [x * 10 for x in range(shape6[0])]),
+        "negative": ("x", [str(-x+x%2) for x in range(shape6[0])]),
+        "nans": ("x", np.array([np.nan] * shape6[0], dtype="float64"))
+    }
+    da6 = xr.DataArray(data=data6, dims=dims, coords=coords6)
+    da6_file = "xarray_align_debug_da6.nc"
+    # da6.to_netcdf(da6_file)
+    # da6_reloaded = xr.open_dataarray(da6_file)
+
+    shape7 = (30, 10)
+    data7 = np.full(shape7, np.nan, dtype="float64")
+    for i in range(shape7[0]):
+        data7[i, i % shape7[1]] = float(-i)
+    coords7 = {
+        "lower": ("y", list(string.ascii_lowercase[shape7[1]:2 * shape7[1]])),
+        "upper": ("y", [c + string.ascii_uppercase for c in list(string.ascii_uppercase[shape7[1]:2 * shape7[1]])]),
+        "tens": ("x", [x * 10 for x in range(shape7[0], 2 * shape7[0])]),
+        "negative": ("x", [str(-x+x%2) for x in range(shape7[0], 2 * shape7[0])]),
+        "nans": ("x", np.array([np.nan] * shape7[0], dtype="float64"))
+    }
+    da7 = xr.DataArray(data=data7, dims=dims, coords=coords7)
+    da7_file = "xarray_align_debug_da7.nc"
+    # da7.to_netcdf(da7_file)
+    # da7_reloaded = xr.open_dataarray(da7_file)
+
+    # for da in (da6_reloaded, da7_reloaded):
+    for da in (da6, da7):
+        da.set_index(append=True, inplace=True, **coords_d)
+    # aligned = xr.align(da6_reloaded, da7_reloaded, join="outer")
+    aligned = xr.align(da6, da7, join="outer")
+    print(aligned)
+    print([np.nonzero(~np.isnan(da)) for da in aligned])
+
+
+def main():
+    # print(xr.show_versions())
+    # align_debug()
+    align_bug_reproduce()
+
+if __name__ == '__main__':
+    main()
+
+

brainscore_vision/data/david2004/test.py

@@ -0,0 +1,8 @@
+import pytest
+
+from brainscore_vision import load_dataset
+
+
+@pytest.mark.private_access
+def test_existence():
+    assert load_dataset('David2004') is not None

brainscore_vision/data/deng2009/__init__.py

@@ -0,0 +1,22 @@
+from brainscore_vision import stimulus_set_registry
+from brainscore_vision.data_helpers.s3 import load_stimulus_set_from_s3
+
+BIBTEX = """@INPROCEEDINGS{5206848,
+  author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Kai Li and Li Fei-Fei},
+  booktitle={2009 IEEE Conference on Computer Vision and Pattern Recognition}, 
+  title={ImageNet: A large-scale hierarchical image database}, 
+  year={2009},
+  volume={},
+  number={},
+  pages={248-255},
+  doi={10.1109/CVPR.2009.5206848}}"""
+
+
+# stimulus set
+stimulus_set_registry['imagenet_val'] = lambda: load_stimulus_set_from_s3(
+    identifier="imagenet_val",
+    bucket="brainio-brainscore",
+    csv_sha1="ff79dcf6b0d115e6e8aa8d0fbba3af11dc649e57",
+    zip_sha1="78172d752d8216a00833cfa34be67c8532ad7330",
+    csv_version_id="7TGln_E5k.V8NdNYGB55mX9JYF0s7R4N",
+    zip_version_id="7DJHWmhFvqTYblo8amGykHR3LYyS.3UC")