Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Updated timeseries generation with GenTS #143

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Updated timeseries generation with GenTS #143

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
4e7753d
Added GenTS to be installed in dev environment
AgentOxygen Oct 9, 2024
d556170
Implemented GenTS using current config.yml format
AgentOxygen Oct 15, 2024
3bcc2cd
Fixed bug with yml config path handling
AgentOxygen Oct 15, 2024
8413466
Fixed precommit errors
AgentOxygen Oct 15, 2024
ec17e2e
Cleaned up blank spaces and trailing commas
AgentOxygen Oct 15, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
142 changes: 55 additions & 87 deletions cupid/run.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,8 +25,11 @@
import os

import click
import gents
import intake
import ploomber
from dask.distributed import Client
from dask.distributed import LocalCluster

import cupid.timeseries
import cupid.util
Expand Down Expand Up @@ -83,110 +86,75 @@ def run(
"glc": landice,
}

# Automatically run all if no components specified

if True not in [atmosphere, ocean, land, seaice, landice]:
all = True
for key in component_options.keys():
component_options[key] = True

#####################################################################
# Managing global parameters

global_params = dict()

if "global_params" in control:
global_params = control["global_params"]

global_params["serial"] = serial

####################################################################

if time_series:
timeseries_params = control["timeseries"]
client = None
if not global_params["serial"]:
cluster = LocalCluster(
n_workers=1,
processes=1,
threads_per_worker=1,
memory_limit="2GB",
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a reason to specify memory_limit rather than letting the LocalCluster object figure it out based on available resources? It seems like it could be problematic (if less than 2 GB / core is available) or unnecessarily restrictive (if more than 2 GB / core is available)

)

# general timeseries arguments for all components
num_procs = timeseries_params["num_procs"]

for component, comp_bool in component_options.items():
if comp_bool:

# set time series input and output directory:
# -----
if isinstance(timeseries_params["case_name"], list):
ts_input_dirs = []
for cname in timeseries_params["case_name"]:
ts_input_dirs.append(global_params["CESM_output_dir"]+"/"+cname+f"/{component}/hist/")
cluster.scale(timeseries_params["num_procs"])
client = Client(cluster)

if "ts_output_dir" not in timeseries_params:
timeseries_params["ts_output_dir"] = "./"

for component in component_options:
if component not in timeseries_params:
continue
include_vars = None
if "vars" in timeseries_params[component]:
include_vars = timeseries_params[component]["vars"]
if include_vars == [] or include_vars == ["process_all"]:
include_vars = None

year_start = None
if "start_years" in timeseries_params[component]:
year_start = timeseries_params[component]["start_years"]
if year_start == []:
year_start = None
else:
ts_input_dirs = [
global_params["CESM_output_dir"] + "/" +
timeseries_params["case_name"] + f"/{component}/hist/",
]

if "ts_output_dir" in timeseries_params:
if isinstance(timeseries_params["ts_output_dir"], list):
ts_output_dirs = []
for ts_outdir in timeseries_params["ts_output_dir"]:
ts_output_dirs.append([
os.path.join(
ts_outdir,
f"{component}", "proc", "tseries",
),
])
else:
ts_output_dirs = [
os.path.join(
timeseries_params["ts_output_dir"],
f"{component}", "proc", "tseries",
),
]
year_start = int(year_start[0])
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The current implementation is that start_years (and end_years) should be lists that are the same length as case_name. This lets us, as an example, compare 60 years of a current run against 100 years of a baseline.


year_end = None
if "end_years" in timeseries_params[component]:
year_end = timeseries_params[component]["end_years"]
if year_end == []:
year_end = None
else:
if isinstance(timeseries_params["case_name"], list):
ts_output_dirs = []
for cname in timeseries_params["case_name"]:
ts_output_dirs.append(
os.path.join(
global_params["CESM_output_dir"],
cname,
f"{component}", "proc", "tseries",
),
)
else:
ts_output_dirs = [
os.path.join(
global_params["CESM_output_dir"],
timeseries_params["case_name"],
f"{component}", "proc", "tseries",
),
]
# -----

# fmt: off
# pylint: disable=line-too-long
cupid.timeseries.create_time_series(
component,
timeseries_params[component]["vars"],
timeseries_params[component]["derive_vars"],
timeseries_params["case_name"],
timeseries_params[component]["hist_str"],
ts_input_dirs,
ts_output_dirs,
# Note that timeseries output will eventually go in
# /glade/derecho/scratch/${USER}/archive/${CASE}/${component}/proc/tseries/
timeseries_params["ts_done"],
timeseries_params["overwrite_ts"],
timeseries_params[component]["start_years"],
timeseries_params[component]["end_years"],
timeseries_params[component]["level"],
num_procs,
serial,
logger,
)
# fmt: on
# pylint: enable=line-too-long
year_end = int(year_end[0])

modb = gents.ModelOutputDatabase(
hf_head_dir=global_params["CESM_output_dir"] + "/" + timeseries_params["case_name"],
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What happens if timeseries_params["case_name"] is a list? As mentioned above, this is the case in our key_metrics example, and we want to loop through each case_name, applying the appropriate start_year and end_year (ts_done and overwrite_ts are both lists of the same length as well, though I'm not 100% clear on why ts_done is in the config file)

ts_head_dir=timeseries_params["ts_output_dir"],
dir_name_swaps={"hist": "proc/tseries"},
dir_exclusions=[comp for comp in component_options if comp != component],
timeseries_year_length=10,
overwrite=timeseries_params["overwrite_ts"][0],
include_variables=include_vars,
compression_level=0,
year_start=year_start,
year_end=year_end,
verbosity_level=1,
)
modb.run(client=client, serial=global_params["serial"])
client.shutdown()

# Grab paths

run_dir = os.path.realpath(os.path.expanduser(control["data_sources"]["run_dir"]))
output_dir = run_dir + "/computed_notebooks/" + control["data_sources"]["sname"]
temp_data_path = run_dir + "/temp_data"
Expand Down
1 change: 1 addition & 0 deletions environments/dev-environment.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,4 +18,5 @@ dependencies:
- pyyaml
- xarray
- pip:
- gents
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to install a static version of gents, or should we add it as a submodule and then use -e ../externals/gents to install it? The latter might be helpful as we tweak the CUPiD API for timeseries generation, especially if it will result in changes to gents as well.

- -e ../