Timeline for Sandbox for Proposed Challenges
Current License: CC BY-SA 4.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 24, 2020 at 18:45 | comment | added | Grimmy | This needs an explanation of how "accuracy" is computed. (And you still need to define "most similar"). | |
| Feb 24, 2020 at 16:59 | history | edited | Revolucion for Monica | CC BY-SA 4.0 |
added 1261 characters in body
|
| Feb 24, 2020 at 16:24 | history | edited | Revolucion for Monica | CC BY-SA 4.0 |
added 14 characters in body
|
| Feb 24, 2020 at 16:19 | comment | added | Revolucion for Monica | @Grimmy Porbably I misunderstood "It looks like your reference code fetches output from an external source" then, my apologizes. | |
| Feb 24, 2020 at 16:11 | comment | added | Grimmy |
External source: yes, that's the way I provided example data like the inputs above. I'm confused. Do you use tfhub.dev/google/universal-sentence-encoder-multilingual/3 only to get example data? It sure doesn't look like that in your code.
|
|
| Feb 24, 2020 at 16:10 | comment | added | Grimmy |
Wait, Google Universal Encoder isn't compulsory? That seems to contradict the most similar element in an array of string with another string according to Google Multilingual Universal Sentence encoder. If GMUSE isn't required, this sentence should be replaced by a proper definition of "most similar element", and GMUSE should only be mentioned in the footnotes.
|
|
| Feb 24, 2020 at 16:05 | comment | added | Revolucion for Monica | @Grimmy Google Universal Encoder: that's not compulsory, but I find it very handy and a good tip to start. External source: yes, that's the way I provided example data like the inputs above. Should I hard code the sources? | |
| Feb 24, 2020 at 15:55 | comment | added | Grimmy | It looks like your reference code fetches output from an external source, which is forbidden by default. Please detail within the question how one would implement the Google Universal Sentence Encoder. (If the challenge actually requires querying tfhub.dev/google/universal-sentence-encoder-multilingual/3, I don't think that's gonna be appropriate for CGCC). | |
| Feb 24, 2020 at 15:19 | comment | added | Revolucion for Monica | @Arnauld Indeed, fixed, thanks! | |
| Feb 24, 2020 at 15:19 | history | edited | Revolucion for Monica | CC BY-SA 4.0 |
added 49 characters in body
|
| Feb 24, 2020 at 15:08 | history | answered | Revolucion for Monica | CC BY-SA 4.0 |