A demonstration of using word embeddings to solve word analogy problems. Also a search program to find new analogies. See https://docs.google.com/document/d/1JXdu2-R01JSFqkxkTK0yvhGIwG6ql76uZ7HUyoQXNEM/edit#data:image/png;base64,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A demonstration of using word embeddings to solve word analogy problems. Also a search program to find new analogies. See https://docs.google.com/document/d/1JXdu2-R01JSFqkxkTK0yvhGIwG6ql76uZ7HUyoQXNEM/edit#
de:für _ = _ bis _ _ 110i
de:wenn _ dann _ sonst _
25500When comparing distances one can save some time by avoiding many calls to square root. See also the 'distance from features ...' block.
Reports distance to the center of the space. See https://en.wikipedia.org/wiki/Norm_(mathematics)#Euclidean_norm
See https://en.wikipedia.org/wiki/Dot_product
See https://en.wikipedia.org/wiki/Cosine_similarity There is a faster JavaScript version of this.
See "closest word to ..." that is much much faster.
candidate featurescandidate distanceEuclidean distance between features using Snap! hyperblocks.
Reports the average value of corresponding items in the two lists.
Reports the average distance from 'word' to each of the 'words'.
Euclidean distance Cosine similarityChinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedish
#11datacountmapmany##1#1data listscount
Reports the average of each element in the list.
Apples function to corresponding items in the 'data lists'.
Reports the distance between the points. Works with any dimensions from 1 to what fits in memory. To see how this can be implemented in a slower way in Snap! open the 'distance from features ... to ...' block
de:kombiniere mit _ die Elemente von _ 1See https://en.wikipedia.org/wiki/Cosine_similarity There is a pure Snap! version of this.
de:für jedes _ von _ _
Language to use for speech input and output only if no language or voice is specified. Can be the language name in English or itself. Or the official code for the language such as fr-FR for French in France. Also used as the default language for the word embedding blocks.
Reports all the words that have a list of features accessible via the "features of" reporter block. Data obtained from https://github.com/facebookresearch/fastText
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish SwedishReports the word that is closest to the features. Uses a distance measure which can be either Eucliean Distance or Cosine similarity. You can choose from among 15 languages or leave it blank for the default language. Optionally one can "watch" it search by adding blocks to the "when word found" slot. The block is passed the best word so far, its distance, and the number of words considered so far.
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish SwedishCosine similarityEuclidean distance Cosine similarityReports a location of a word embedding use t-SNE. See https://en.wikipedia.org/wiki/T-distributed_stochastic_neighbor_embedding
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedishstage sizeStage size11122222Convert from 0 to 1 coordinates to Snap! coordinations.
0.30.70.6
LABEL will stamp text on the stage at the given font size. The direction of the text is the direction the sprite is facing, and color will match the pen color.
de:drucke _ in Größe _ Hello!12
Displays the word near similar words.
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedish121Reports the word that is closest to the features. You can choose from among 15 languages or leave it blank for the default language. There is a full-featured version of this block as well.
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish SwedishReports a list of 300 numbers that "describes" the word. You can choose among 15 languages or leave it blank and use the default language. To learn about these numbers see https://en.wikipedia.org/wiki/Word_embedding The new name for this block is "features of word ..."
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedish
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Portuguese Sinhalese Swedish
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Portuguese Sinhalese Swedish
Reports all the words sorted by closeness to 'features'. Optionally can report pairs of words and distances. Distances are measured using cosine similarity.
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedishfalsewords
record_callbacks(finished_callback);falseFinds the average of the features of each word in 'words'.
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish SwedishReports a list created by applying reporter to corresponding elements of the lists.
11Passes a list of lists of 512 features of input 'sentences' to 'process result'.
Reports 512 features for the sentence.
list of features listPasses to 'process result' a tokenization of the sentence.
Reports tokens for the sentence.
tokensReports a list of 512 features for each sentence.
list of features listReports a list of 300 numbers that "describes" the word. You can choose among 15 languages or leave it blank and use the default language. To learn about these numbers see https://en.wikipedia.org/wiki/Word_embedding
Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedish
Uses 'passage' to answer 'question'. The 'answer question' reporter is easier to use and relies upon this block.
This uses BERT to answer questions. See https://blog.tensorflow.org/2020/03/exploring-helpful-uses-for-bert-in-your-browser-tensorflow-js.htmlReports the top answer to the 'question' using the 'passage'.
Reports a list of 5 elements where each element is a list of an answer to the 'question' and its score of how good an answer it is. The highest scoring answer is first.
answers
Will display 'message' in a dialog box with 'title'. User needs to click 'OK' to remove it.
Returns only after eCraft2Learn library is loaded
Should only be called by Load eCraft2Learn library
A concise shortcut to the 'features of ...' block. It uses English.
A concise shortcut to the 'closest word to ...' block. Uses English and excludes no words.
Allows reading global settings programmatically. Eisenberg's Law: Anything you can do from the user interface you should be able to do in your program, and vice versa. This library is just a beginning; there are many UI controls outside of the Settings menu.
pt:o valor da configuração _ Project notes Project name User Presentation mode Language Zoom blocks Stage size Stage scale Retina display support Long form input dialog Plain prototype labels Input sliders Execute on slider change Clicking sound Turbo mode Flat design Keyboard editing Visible stepping Thread safe scripts Prefer smooth animations Flat line ends Codification support Inheritance support Hyper blocks support Visible paletteAllows changing global settings programmatically. This block is for Boolean (checkbox) settings; use SET VALUE for numeric or text values. Eisenberg's Law: Anything you can do from the user interface you should be able to do in your program, and vice versa. This library is just a beginning; there are many UI controls outside of the Settings menu.
pt:altera o valor da configuração _ para _ Presentation mode Retina display support Long form input dialog Plain prototype labels Input sliders Execute on slider change Clicking sound Turbo mode Flat design Keyboard editing Visible stepping Thread safe scripts Prefer smooth animations Flat line ends Codification support Inheritance support Hyper blocks supportAllows changing global settings programmatically. This block is for numeric or text settings; use SET FLAG for Boolean (checkbox) values. Eisenberg's Law: Anything you can do from the user interface you should be able to do in your program, and vice versa. This library is just a beginning; there are many UI controls outside of the Settings menu.
pt:altera o valor da configuração _ para _ Project notes Project name Language Zoom blocks Stage size Stage scale Visible paletteReports a new list containing the items of the input list, but in the opposite order.
ca:capgira _ 1inputresultReports a new list whose items are the same as in the input list, except that if two or more equal items appear in the input list, only the last one is kept in the result.
ca:elimina els duplicats de _ 1Reports a sorted version of the list in its first input slot, using the comparison function in the second input slot. For a list of numbers, using < as the comparison function will sort from low to high; using > will sort from high to low.
ca:ordena _ segons criteri _ copy of datasplitmerge11#1#2The second input is an "association list," a list of two-item lists. Each of those smaller lists has a "key" as its first item and a "value" as its second. ASSOC reports the first key-value pair in the association list whose key matches the first input.
ca:associació _ _ 11This block carries out the given script for each item of the given list, like the primitive FOR EACH. What's different is that it provides the # variable, which will contain the item number in the list of each item in turn, 1 while processing item 1, and so on.
ca:_ per cada _ de _ _ Takes as input a function of N inputs and N lists. The function is called with item 1 of all the lists as its inputs, with item 2 of all the lists as its inputs, and so on. (The lists should all be the same length.)
ZIP takes any number of lists as inputs. The lists should all be the same length. ZIP reports a list of lists in which the first item is a list of all the first items, the second item is a list of all the second items, etc. Viewing the inputs as the rows of a matrix, ZIP reports its transpose.
1The identity function reports its input.
Finds the distances between pairs taken from the incoming list of 4 words and if less than threshold reports them.
2Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedishpossibilities left to explore-2An alternative using Euclidean distance is to use cosine similarity. Try replacing the distance calculations with the following.Reports all analogies with a distance score less than 'threshold' taken from the 'number of words' closest to 'word'.
102Chinese English Finnish French German Greek Hindi Indonesian Italian Japanese Lithuanian Portuguese Sinhalese Spanish Swedishbest analogies11Sort the analogies by distnace between the pairs.Click on https://en.wikipedia.org/wiki/Combination
data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAeAAAAFoCAYAAACPNyggAAAAAXNSR0IArs4c6QAADoVJREFUeF7t1cEJAAAIxDDdf2m3sJ+4wEEQuuMIECBAgACBd4F9XzRIgAABAgQIjAB7AgIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECAiwHyBAgAABAoGAAAfoJgkQIECAgAD7AQIECBAgEAgIcIBukgABAgQICLAfIECAAAECgYAAB+gmCRAgQICAAPsBAgQIECAQCAhwgG6SAAECBAgIsB8gQIAAAQKBgAAH6CYJECBAgIAA+wECBAgQIBAICHCAbpIAAQIECByxcQFpoRMBzwAAAABJRU5ErkJggg==Visualizer of word cloud https://projector.tensorflow.org/?config=https://ecraft2learn.github.io/ai/word-embeddings/en/projector.jsonFinding out what word is closest to (king+(woman-man))A good analogy has a short distance from A-B to C-D.0.941176550919114,woman,man,women,men0.941176550919114,woman,women,man,men1.257371579923771,women,men,husband,wife1.257371579923771,women,husband,men,wife1.3622084311881206,woman,husband,man,wife1.3622084311881206,woman,man,husband,wife1.3833505376440198,woman,lady,women,ladies1.3833505376440198,woman,women,lady,ladies1.4081962682808096,women,wife,men,husband1.4388500373562219,woman,lady,men,ladies1.4388500373562219,woman,men,lady,ladies1.4396605745799946,man,men,lady,ladies1.4396605745799946,man,lady,men,ladies1.4566445448358356,lady,man,women,men1.4566445448358358,man,men,lady,women1.5432348168700698,man,men,girl,women1.54323481687007,girl,man,women,men1.545612402900546,woman,husband,lady,wife1.545612402900546,woman,lady,husband,wife1.5643463587070479,men,man,women,person1.5668200119988251,woman,lady,women,men1.5668200119988251,woman,women,lady,men1.576054558065805,woman,person,women,men1.576054558065805,women,woman,men,person1.6143563361290474,woman,girl,women,men1.6143563361290474,woman,women,girl,men1.6250511038118156,woman,wife,man,husband1.6286750780926191,man,wife,men,husband1.637632281679865,women,lady,men,ladies1.637632281679865,lady,ladies,women,men1.650360890835699,woman,husband,girl,wife1.650360890835699,woman,girl,husband,wife1.671606661269331,woman,men,lady,women1.6766730599613036,woman,wife,women,husband1.6811116441212344,man,wife,person,husband1.703950938260841,husband,lady,wife,ladies1.703950938260841,lady,ladies,husband,wife1.7134204416896632,lady,women,girl,men1.7134204416896632,lady,girl,women,men1.7277494841556162,ladies,lady,men,person1.7277494841556162,person,lady,men,ladies1.7313772177084927,woman,husband,women,wife1.7313772177084927,woman,women,husband,wife1.7383900454155852,woman,lady,man,ladies1.7383900454155852,woman,man,lady,ladies1.751438066275825,woman,wife,lady,husband1.7554407081983712,wife,lady,husband,ladies1.7569146478984117,lady,wife,girl,husband1.7586774235203002,lady,person,women,men1.7586774235203004,women,lady,men,person1.7595232791867228,men,lady,women,ladies1.7613124765356092,men,man,women,wife1.7613124765356092,wife,man,women,men1.7634131932136603,women,ladies,husband,wife1.7634131932136607,husband,women,wife,ladies1.7777295660476597,woman,men,husband,wife1.7777295660476597,woman,husband,men,wife1.7795143747663296,lady,men,girl,women1.7817829244888386,man,lady,women,ladies1.7817829244888388,man,women,lady,ladies1.7822323249228755,woman,men,girl,women1.7828121774320476,woman,person,husband,wife1.7828121774320478,woman,husband,person,wife1.78723138960796,man,person,husband,wife1.78723138960796,man,husband,person,wife1.7907748602211278,women,woman,men,wife1.7907748602211278,woman,wife,women,men1.795038403488905,man,husband,men,wife1.795038403488905,man,men,husband,wife1.7952691051761567,lady,girl,husband,wife1.7952691051761567,lady,husband,girl,wife1.796040467250112,woman,man,lady,person1.796040467250112,woman,lady,man,person1.8058927709030785,man,wife,lady,husband1.8059262637217501,men,man,women,husband1.8059262637217501,husband,man,women,men1.8072262143959736,woman,wife,girl,husband1.8159299518428564,men,ladies,husband,wife1.8159299518428564,husband,men,wife,ladies1.8218114556671337,woman,women,man,ladies1.8218114556671339,woman,man,women,ladies1.8268708575047112,man,wife,girl,husband1.8293857630363257,women,man,husband,wife1.8293857630363257,man,wife,women,husband1.834637198467316,man,person,women,men1.8346371984673162,women,man,men,person1.835159497155492,man,husband,lady,wife1.8351594971554923,man,lady,husband,wife1.837244248868397,women,woman,men,husband1.837244248868397,woman,husband,women,men1.8373680442415457,woman,wife,men,husband1.842584893566644,woman,lady,man,girl1.842584893566644,woman,man,lady,girl1.843034047433741,man,woman,men,ladies1.843034047433741,woman,ladies,man,men1.8448117464933922,woman,men,man,women1.8514979179032311,woman,girl,women,ladies1.8514979179032311,woman,women,girl,ladies1.8527389535495815,lady,wife,women,men1.8527389535495817,women,lady,men,wife1.8564306504687964,ladies,lady,men,husband1.8564306504687966,husband,lady,men,ladies1.8584173858420503,men,man,women,ladies1.8584173858420503,ladies,man,women,men1.8601133917049253,ladies,lady,men,wife1.8601133917049255,wife,lady,men,ladies1.8772982368286606,lady,ladies,girl,men1.8772982368286606,girl,lady,men,ladies1.8821108176725405,girl,person,women,men1.8821108176725407,women,girl,men,person1.8912569999870457,ladies,lady,women,husband1.8912569999870457,husband,lady,women,ladies1.8921195628183751,man,men,girl,ladies1.8921195628183753,man,girl,men,ladies1.8921577339112088,man,husband,girl,wife1.8921577339112088,man,girl,husband,wife1.8961906180550525,woman,wife,person,husband1.89688818858677,ladies,man,lady,person1.89688818858677,person,man,lady,ladies1.9047018244334202,ladies,lady,women,person1.9047018244334202,person,lady,women,ladies1.907816765310548,woman,girl,man,lady1.9103603141815955,men,woman,man,lady1.9112092480939913,lady,ladies,girl,women1.9112092480939913,girl,lady,women,ladies1.9188663762753257,woman,girl,man,person1.9188663762753257,woman,man,girl,person1.920459473667695,woman,ladies,man,lady1.922383008143799,lady,wife,person,husband1.9235660009472,woman,girl,men,ladies1.9235660009472,woman,men,girl,ladies1.925052321366876,women,woman,men,ladies1.925052321366876,woman,ladies,women,men1.9385042919735822,woman,ladies,husband,wife1.9385042919735822,husband,woman,wife,ladies1.9471920372680245,ladies,lady,women,wife1.9471920372680245,wife,lady,women,ladies1.951616752848776,woman,man,lady,wife1.951616752848776,woman,lady,man,wife1.9591877960011892,women,lady,husband,wife1.9591877960011894,lady,wife,women,husband1.9652505311028408,woman,man,women,person1.9652505311028408,woman,women,man,person1.9774889658352082,woman,man,lady,men1.9774889658352082,woman,lady,man,men1.9846268037089498,wife,men,husband,ladies1.9850799127491061,women,lady,men,husband1.9850799127491063,lady,husband,women,men1.9853057069378515,woman,girl,lady,ladies1.9853057069378515,woman,lady,girl,ladies1.987401426486355,men,woman,women,husband1.9892346342249323,lady,person,husband,wife1.9892346342249323,lady,husband,person,wife1.9908895298333353,men,woman,man,girl1.9908895298333353,girl,woman,man,men0