Closed superliuxz closed 5 years ago
could you explain what scale
key is?
could you explain what
scale
key is?
Yes. scale
~= height == the # of ratings (numRatings
key).
So why don't I just use the # of ratings as the height of each bar?
Say we have two genres, Action
and Drama
, Action
is a much more popular genre than Drama
. What could end up happening is that: under Drama
, we have two movies (d1
and d2
), with 10000 and 20000 ratings, and under Action
, we have two movies (a1
and a2
) with 100 and 200 ratings.
Imagine that we plot the grouped bar chart on d1
, d2
, a1
and a2
. The height of d1
and d2
would be 100 times higher than a1
and a2
.
Hence I provide the normalized height - scale
, each bar height is normalized with respect to the total number of ratings within the belong genre. By doing so, when plotted, all the bars on the graph will have more or less the same height.
What I would want to plot is that, on the vertical axis there is no units. Instead, when the mouse hovers over the bar, we can use a tooltip to display the numRatings
key.
I hope this makes sense.
What I am thinking is to have a bar chart, with the horizontal axis being the 18 different genres, and in each genre we plot the top 5 most rated movies of all time. The bar hight is the
scale
key of each movie, the name of the bar can be themovie
-year
, and thenumRatings
andavgRating
can be used as tooltips to display more data. The plot will be very similar to the existing plot # 3.@soroushysfi , is this possible?
Can you elaborate more? I didn't get what exactly you mean. can you show it in a sample bar chart?(a simple sketch would be fine).
and I'm getting this error when I call the API:
Error: no such table: rating_stats
Is it running on 27m dataset?
What I am thinking is to have a bar chart, with the horizontal axis being the 18 different genres, and in each genre we plot the top 5 most rated movies of all time. The bar hight is the
scale
key of each movie, the name of the bar can be themovie
-year
, and thenumRatings
andavgRating
can be used as tooltips to display more data. The plot will be very similar to the existing plot # 3. @soroushysfi , is this possible?Can you elaborate more? I didn't get what exactly you mean. can you show it in a sample bar chart?(a simple sketch would be fine). and I'm getting this error when I call the API:
Error: no such table: rating_stats
Is it running on 27m dataset?
Error: no such table: rating_stats
: checkout latest master, then remove assignment1.db
, then python3 preproc.py
.
Please see the following:
What I am thinking is to have a bar chart, with the horizontal axis being the 18 different genres, and in each genre we plot the top 5 most rated movies of all time. The bar hight is the
scale
key of each movie, the name of the bar can be themovie
-year
, and thenumRatings
andavgRating
can be used as tooltips to display more data. The plot will be very similar to the existing plot # 3. @soroushysfi , is this possible?Can you elaborate more? I didn't get what exactly you mean. can you show it in a sample bar chart?(a simple sketch would be fine). and I'm getting this error when I call the API:
Error: no such table: rating_stats
Is it running on 27m dataset?
Error: no such table: rating_stats
: checkout latest master, then removeassignment1.db
, thenpython3 preproc.py
.Please see the following:
Sure! it will be easy.
It sais here that this branch has some conflicts on the app.py file. Is it ok to merge?
It sais here that this branch has some conflicts on the app.py file. Is it ok to merge?
I will have to resolve the conflicts first. Will let you know when I resolve them, then you can merge. Thanks.
It sais here that this branch has some conflicts on the app.py file. Is it ok to merge?
Done.
What I am thinking is to have a bar chart, with the horizontal axis being the 18 different genres, and in each genre we plot the top 5 most rated movies of all time. The bar hight is the
scale
key of each movie, the name of the bar can be themovie
-year
, and thenumRatings
andavgRating
can be used as tooltips to display more data. The plot will be very similar to the existing plot # 3.@soroushysfi , is this possible?