Make the clusterPlots notebook consistent with telemetry schema #68
Conversation
poojanilangekar
force-pushed
the
cluster-plots
branch
from
de87352 to
6f7e7fa
Compare
poojanilangekar
force-pushed
the
cluster-plots
branch
from
6f7e7fa to
f574af1
Compare
| @@ -50,7 +50,9 @@ | |||
| "# --- Data manipulations --- #\n", | |||
| "\n", | |||
| "# Retrieve relevant data from duckdb.\n", | |||
| "EXP_DATA = retrieve_experiment_df(con, EXPERIMENT_ID, EXPERIMENT_START_TIME)\n", | |||
| "EXP_DATA = pd.DataFrame()\n", | |||
There was a problem hiding this comment.
For cluster plots, we only plot one experiment at a time, thus we only have one experiment id (EXPERIMENT_ID = "" instead of EXPERIMENT_ID = [""] like in the other notebooks), so you wouldn't actually need the for-loop here. The reason is that you're plotting different phases as different lines in the plot (vs different experiments as different lines in the plot which is what we do in the latency plots), thus you can actually only plot one experiment at a time. If you have any suggestions on how to visualize multiple experiments in the same plot, let me know! :)
|
Hi @poojanilangekar, thanks for the PR and addressing my review comments. |
jcamachor
force-pushed
the
main
branch
6 times, most recently
from
44a31f1 to
13a6b2c
Compare
jcamachor
force-pushed
the
main
branch
2 times, most recently
from
868d4f8 to
ff310c5
Compare
jcamachor
force-pushed
the
main
branch
3 times, most recently
from
4def705 to
7876be3
Compare
Making the clusterPlots notebook consistent with execTimePlots and storagePlots notebooks.