Mirrors/the-algorithm-ml
2
0
Fork 0
mirror of https://github.com/twitter/the-algorithm-ml.git synced 2024-12-23 06:41:49 +01:00
Code Issues Packages Projects Releases Wiki Activity

Reformat README.md

Browse Source 
Make this file more readable
This commit is contained in:
yang1fan2 2023-04-01 15:01:42 -07:00 committed by GitHub
parent 78c3235eee
commit d81111fc59
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 24 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
projects/home/recap/README.md
View File

@ -6,28 +6,28 @@ The model receives various features, describing the Tweet and the user whose tim
Those are: Those are:
"recap.engagement.is_favorited": The probability the user will favorite the Tweet. - "recap.engagement.is_favorited": The probability the user will favorite the Tweet.
"recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": The probability the user will click into the conversation of this Tweet and reply or Like a Tweet. - "recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": The probability the user will click into the conversation of this Tweet and reply or Like a Tweet.
"recap.engagement.is_good_clicked_convo_desc_v2": The probability the user will click into the conversation of this Tweet and stay there for at least 2 minutes. - "recap.engagement.is_good_clicked_convo_desc_v2": The probability the user will click into the conversation of this Tweet and stay there for at least 2 minutes.
"recap.engagement.is_negative_feedback_v2": The probability the user will react negatively (requesting "show less often" on the Tweet or author, block or mute the Tweet author) - "recap.engagement.is_negative_feedback_v2": The probability the user will react negatively (requesting "show less often" on the Tweet or author, block or mute the Tweet author)
"recap.engagement.is_profile_clicked_and_profile_engaged": The probability the user opens the Tweet author profile and Likes or replies to a Tweet. - "recap.engagement.is_profile_clicked_and_profile_engaged": The probability the user opens the Tweet author profile and Likes or replies to a Tweet.
"recap.engagement.is_replied": The probability the user replies to the Tweet. - "recap.engagement.is_replied": The probability the user replies to the Tweet.
"recap.engagement.is_replied_reply_engaged_by_author": The probability the user replies to the Tweet and this reply is engaged by the Tweet author. - "recap.engagement.is_replied_reply_engaged_by_author": The probability the user replies to the Tweet and this reply is engaged by the Tweet author.
"recap.engagement.is_report_tweet_clicked": The probability the user will click Report Tweet. - "recap.engagement.is_report_tweet_clicked": The probability the user will click Report Tweet.
"recap.engagement.is_retweeted": The probability the user will ReTweet the Tweet. - "recap.engagement.is_retweeted": The probability the user will ReTweet the Tweet.
"recap.engagement.is_video_playback_50": The probability (for a video Tweet) that the user will watch at least half of the video - "recap.engagement.is_video_playback_50": The probability (for a video Tweet) that the user will watch at least half of the video
For ranking the candidates these predictions are combined into a score by weighting them: For ranking the candidates these predictions are combined into a score by weighting them:
"recap.engagement.is_favorited": 0.5 - "recap.engagement.is_favorited": 0.5
"recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": 11* (the maximum prediction from these two "good click" features is used and weighted by 11, the other prediction is ignored). - "recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": 11* (the maximum prediction from these two "good click" features is used and weighted by 11, the other prediction is ignored).
"recap.engagement.is_good_clicked_convo_desc_v2": 11* - "recap.engagement.is_good_clicked_convo_desc_v2": 11*
"recap.engagement.is_negative_feedback_v2": -74 - "recap.engagement.is_negative_feedback_v2": -74
"recap.engagement.is_profile_clicked_and_profile_engaged": 12 - "recap.engagement.is_profile_clicked_and_profile_engaged": 12
"recap.engagement.is_replied": 27 - "recap.engagement.is_replied": 27
"recap.engagement.is_replied_reply_engaged_by_author": 75 - "recap.engagement.is_replied_reply_engaged_by_author": 75
"recap.engagement.is_report_tweet_clicked": -369 - "recap.engagement.is_report_tweet_clicked": -369
"recap.engagement.is_retweeted": 1 - "recap.engagement.is_retweeted": 1
"recap.engagement.is_video_playback_50": 0.005 - "recap.engagement.is_video_playback_50": 0.005
We cannot release the real training data due to privacy restrictions. However, we have included a script to generate random data to ensure you can run the model training code. We cannot release the real training data due to privacy restrictions. However, we have included a script to generate random data to ensure you can run the model training code.
@ -35,11 +35,14 @@ We cannot release the real training data due to privacy restrictions. However, w
To try training the model (assuming you have already followed the repo setup instructions and are inside a virtualenv). To try training the model (assuming you have already followed the repo setup instructions and are inside a virtualenv).
Run Run
```
$ ./projects/home/recap/scripts/create_random_data.sh $ ./projects/home/recap/scripts/create_random_data.sh
```
This will create some random data (in $HOME/tmp/recap_local_random_data). This will create some random data (in $HOME/tmp/recap_local_random_data).
```
$ ./projects/home/recap/scripts/run_local.sh $ ./projects/home/recap/scripts/run_local.sh
```
This will train the model (for a small number of iterations). Checkpoints and logs will be written to $HOME/tmp/runs/recap_local_debug. This will train the model (for a small number of iterations). Checkpoints and logs will be written to $HOME/tmp/runs/recap_local_debug.