plot feedback - Githubissues

To. Oriol

structure

Starting with 69 .csvs, (3(counterfactual of vaccine amount) *23(parameters) for sensitivity calculation), I store them and plot:

Writing test codes

        vv["x1_pcth1"] = vv["x1_pcsth1"].sum(dim="space") # TODO: compare same value 
        vv["x2_pnth1"] = vv["x2_pnsth1"].sum(dim="space")
        # print("VERIFY1: vensim-xarray space aggregation is same")
        # print("VENSIM :", paramvv["Cost of Deaths Cumulative All"].values.flatten())
        # print("xarray :", vv["x1_pcth1"].values)

dims, coords

how do you usually think coords? I noticed all my coords have one dims and wish to know how to use this more flexibly
how does coords, dims change throughout the plotting process? Ari thinks: dims should not change but I might have multiple coords
misc
directory No such file or directory: 'V79/CSV/Covidparam_Variant Intro Start Time_1.01.csv'
module ModuleNotFoundError: No module named 'VST' + file structures in general (code, input csv_data, nc_data)

code

# x
x1 = ["cost"]
x2 = ["population"]

# function
f1x = ["mvv", "avv"] #better if functional form
f2x = ["mvv_sens", "avv_sens"]

# param
t_sharing_sbatched_params, ts_sharing_params  = gen_param_names()[1:]
param_lst = t_sharing_sbatched_params + ts_sharing_params + ["baseline"]

# component_time_vec
costcomp_tv = ['death', 'gdp', 'hospitalization']
noncostcomp_tv = ['population']
# space
state_names  = gen_param_names()[0]

# time
time = pd.date_range(start='10/15/2019', periods=1071)

# hyperparameter/action/counterfactual
vax_amt_lst = [1.01, 1, 0]
vax_time_lst = [range(-50, 60, 10)]

dv = {
    "dims": {"x1", "x2", "f1x", "f2x", "param", "costcomp_tv", "noncostcomp_tv", "space", "time", "hp_vax_amt", "hp_vax_time"}, #action as hparam
    "coords": { #pcstvh
        'x1': {"dims": ("x1"), "data": x1}, #COST ($) TODO: if I use ["cost"], should i change x1_pcsth1 as "x1" or "cost"?
        'x2': {"dims": ("x2"), "data": x2}, #POPULATION (person)

        'f1x': {"dims": ("f1x"), "data": f1x}, #MARGINAL AVERAGE VACCINE VALUE PER PERSON ($/person)
        'f2x': {"dims": ("f2x"), "data": f2x}, #MARGINAL AVERAGE VACCINE VALUE PER PERSON w.r.t. PARAMETER ($/person/punit)

        'param': {"dims": ("param"), "data": param_lst},

        'costcomp_tv': {"dims": ("costcomp_tv"), "data": costcomp_tv},
        'noncostcomp_tv': {"dims": ("noncostcomp_tv"), "data": noncostcomp_tv},

        'space': {"dims": ("space"), "data": state_names},
        'time': {"dims": ("time"), "data": time},
        'hp_vax_amt': {"dims": ("hp_vax_amt"), "data": [str(va) for va in vax_amt_lst]},
        'hp_vax_time': {"dims": ("hp_vax_time"), "data": [str(vt) for vt in vax_time_lst]}
    },
    "data_vars": {  # 5+3 three vdot_ coordinate is only needed for data variables with 'p' coordinate
        "x1_pcsth1": {"dims": ("x1", "param", "costcomp_tv", "space", "time", "hp_vax_amt"), "data": np.zeros(shape=(len(x1), len(param_lst), len(costcomp_tv), len(state_names), len(time), len(vax_amt_lst)))}, # vaccine value for each component, space, time, for vaxinc 0, 1, 1.01
        "x2_pnsth1": {"dims": ("x2", "param", "noncostcomp_tv", "space", "time", "hp_vax_amt"), "data": np.zeros(shape=(len(x2), len(param_lst), len(noncostcomp_tv), len(state_names), len(time), len(vax_amt_lst)))},

        "x1_cth2": {"dims": ("x1","costcomp_tv", "time", "hp_vax_time"), "data": np.zeros(shape=(len(x1),len(costcomp_tv), len(time), len(vax_time_lst)))},

        "f1x_pcsth1": {"dims": ("f1x", "param", "costcomp_tv", "space", "time","hp_vax_amt"), "data": np.zeros(shape=(len(f1x), len(param_lst), len(costcomp_tv),  len(state_names), len(time), len(vax_amt_lst)))}, #marginal vaccine value per person
        "f2x_pcst": {"dims": ("f2x", "param", "costcomp_tv", "space", "time",), "data": np.zeros(shape=(len(f2x), len(param_lst), len(costcomp_tv),  len(state_names), len(time)))}, #marginal vaccine value per person w.r.t. parameter
    },
    "attrs": {"title": "1.vaccine value disaggregated by parameter, time, space, component, 2.counterfactual by parameter for validity, 3.counterfactual by datahp for summary stat."}
    # 1 and 2 are descriptive, 3 is prescriptive
}
vv = xr.Dataset.from_dict(dv)

hyunjimoon / VaccineMisinf

plot feedback #15

structure

Writing test codes

dims, coords

misc

code