I am looking for some help with finding a list of categories recognized by Google Vision API either by Object Localizer API or by Product Search API.
I searched the issue and found an old post here https://groups.google.com/g/cloud-vision-discuss/c/Gpo2h-Rw7gA/m/uSHH-r8lAgAJ and linked StackOverflow questions.
However, after a thorough examination of Open Images V6 and Google Knowledge Graph, I still do not see a clear connection between Open Images category tree and Vision API Product Search.
Product Search API has 5 main productCategories: homegoods, apparel, toys, packedgoods, general. (link: https://cloud.google.com/vision/product-search/docs/product-categories)
However, Open Images does not have any of these categories https://storage.googleapis.com/openimages/2018_04/bbox_labels_600_hierarchy_visualizer/circle.html.
Can anyone help me with a list of detailed categories returned by a detector in Product Search and corresponding top-category (homegoods, apparel, toys, packedgoods, general)?
I would like what categories can be detected in each of the top-category, so for example: "Apparel": ["Pants", "Skirt", "Shoes", ...].
Thanks in advance.
Related
When I look in label_annotions of the Google Vision API, the "score" and "topicality" field values are always the same. This is also for example the case here. According to this reference topicality refers to "the relevancy of the ICA (Image Content Annotation) label to the image" whereas score has replaced "confidence". Though it's now not so clear to me what "score" actually means.
Are these supposed to be always the same? What does that mean?
As you well mentioned, these "score" and "topicality" fields always display the same value. Based on my understanding, this is an existing bug on the Vision API functionality that needs to be fixed in a future release. I recommend you to take a look on this ticket that has been created to review this issue, as well as the Release Notes documentation of Vision API to keep the track of the new functionalities added to the service.
This question already has answers here:
How to get place description from Google Places API
(2 answers)
Closed 3 years ago.
I'd like to understand how or if it is possible to retrieve the same business category value via the Google Places API as the one used next to the place listing in Google search results. While the API does return a list of business categories and much other information beside, the value used on their search results page is (a) singular and (b) often in much better definition than those returned in the API response. Examples as follows; any advice as to its availability much appreciated.
Name: Sydney Showboats
Google Search
Url: https://www.google.co.uk/search?q=Sydney+Showboats
Response Category:
Cruise agency
Google Places API:
Url: https://maps.googleapis.com/maps/api/place/nearbysearch/json?location=-33.8670522,151.1957362&radius=500&type=restaurant&keyword=cruise&key=YOUR_API_KEY
Response Categories:
["travel_agency",
"restaurant",
"food",
"point_of_interest",
"establishment"
]
NB: The same is also true for the response from the Google Places Detail API.
How can I retrieve the finer, single term Cruise agency rather than the higher-level list of categories returned via the API?
I'm still experimenting with all the API calls, but I switched from nearby search to Text Search (https://developers.google.com/places/web-service/search).
Text Search, with query param, started showing me results that matched what was on maps.google.com.
query — The text string on which to search, for example: "restaurant".
The Google Places service will return candidate matches based on this
string and order the results based on their perceived relevance. This
parameter becomes optional if the type parameter is also used in the
search request.
I want get all wikipedia categorys name with api.
i search it and find some way like dumb data or sql data but i want make it this with wikipedia api. Also i found it some great link for wikipedia tools like wmflabs.org but this tools not have my need.
i want use wikipedia api and result json data and get english wikipedia category name list. All categories content is here-> https://en.wikipedia.org/wiki/Portal:Contents/Categories
i need same style category name lis data all other languages.
i find it some query but not enough like(find subcategorys in weapons category)-> https://en.wikipedia.org/w/api.php ?action=query&list=categorymembers&cmtitle=Category:Weapons
Is any solution have for this situation?
All the categories belong to the Category namespace, that has id=14, so you can use allpages API and filter by namespace using the apnamespace parameter. Here's an example query.
I have recently added an Amazon Search to my website and am having an issue getting the full response I need. In around 10% of my calls I get zero hits even though Amazon's site itself display numerous hits. Here is my call to their service:
results = Amazon::Ecs.item_search(query_text, {
:search_index => 'All',
:response_group => 'ItemAttributes, Images, Offers',
:item_page=>’1'
})
90% of the time the results I get back are just fine, but with any of the following query texts this search returns 0 hits:
hoover bag 440004496 type S
Citizen Women’s Stainless Steel Bracelet Watch 18mm EJ5850-57E
mope space pack iPhone 5
Yet when I go to Amazon.com directly and search on any of these targets I do get plenty of relevant responses.
What happens is Amazon displays, for instance, "Your search "hoover bag 440004496 type S" did not match any products" and then goes on to display "hoover bag 440004496 type S” See all 322 results…” and plenty of good, relevant product matches that I would love to display on my page.
You can see the results (or lack thereof) at my site, www.FastForward.menu
How do I get these helpful responses through the API? Preferably in the same format that any populated initial item_search response is in? Preferably in the same call as the initial response so I don’t have to write more parsing code?
There is a “Similar products” API call, but to do that you need at least one product to pass in, which I do not have. See here: http://docs.aws.amazon.com/AWSECommerceService/latest/DG/SimilarityLookup.html
How do I get this list of “similar” products through the API so that my Amazon searched don’t come up completely empty around 10% of the time?
I am sure others have run into this same question.
Please feel free to ask questions or to make suggestions as to how I can improve this question.
I am trying to figure out the right parameters for ItemSearch such that the API call will return the same result as on the website. Currently I am using these params it is not consistent with the website.
url_params = dict(
Service='AWSECommerceService',
Operation='ItemSearch',
AssociateTag=AMAZON_ASSOCIATETAG,
SearchIndex='All',
AWSAccessKeyId=AWS_ACCESS_KEY_ID,
Keywords=keywords,
ResponseGroup='Small,OfferSummary,Images'
)
For example if 'iphone%205s' is put in keywords, the API returns a list of iPhone protectors, while searching on the website gives iPhones as top results.
I am also trying to figure out why this is for book searches. One possible answer I found online was that the websites search feature might use more elaborate queries than just one simple API call. For example, it may take into account other factors into the search (not sure what that might be but it may).