def test_jsondump(self):
dict_list = {'a': [1.0, 2.0, 3.0]}
file_list = os.path.join(TMPDIR, 'list.json')
jsondump(file_list, dict_list)
with open(file_list) as fd:
self.assertEqual(json.load(fd), dict_list)
dict_vec = {'a': np.repeat(3, 4)}
file_vec = os.path.join(TMPDIR, 'vec.json')
jsondump(file_vec, dict_vec)
with open(file_vec) as fd:
self.assertEqual(json.load(fd), dict_vec)
dict_zero_vec = {'a': []}
file_zero_vec = os.path.join(TMPDIR, 'empty_vec.json')
jsondump(file_zero_vec, dict_zero_vec)
with open(file_zero_vec) as fd:
self.assertEqual(json.load(fd), dict_zero_vec)
dict_zero_matrix = {'a': [[], [], []]}
file_zero_matrix = os.path.join(TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
self.assertEqual(json.load(fd), dict_zero_matrix)
arr = np.zeros(shape=(5, 0))
dict_zero_matrix = {'a': arr}
file_zero_matrix = os.path.join(TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
self.assertEqual(json.load(fd), dict_zero_matrix)
def main():
here = os.path.dirname(os.path.realpath(__file__))
input_path_tecrdb = os.path.join(here, INPUT_PATH_TECRDB)
input_path_compounds = os.path.join(here, INPUT_PATH_COMPOUNDS)
input_path_microspecies = os.path.join(here, INPUT_PATH_MICROSPECIES)
tecrdb = pd.read_csv(input_path_tecrdb, index_col=0).loc[filter_tecrdb]
stoichiometry = get_stoichiometry(tecrdb)
cids = stoichiometry['compound_id'].unique()
rids = stoichiometry['reaction_id'].unique()
compounds = (
pd.read_csv(input_path_compounds, index_col=0)
.loc[lambda df: df['compound_id'].isin(cids)]
)
microspecies = (
pd.read_csv(input_path_microspecies, index_col=0)
.loc[lambda df: df['compound_id'].isin(cids)]
)
pkas, pkmgs = (
microspecies
.groupby(['compound_id', n])[[pk, "name"]].first()
.dropna()
.reset_index()
for n, pk in [('nh', 'pka'), ('nmg', 'pkmg')]
)
stan_codes = get_stan_codes(stoichiometry)
stan_input = get_stan_input(
tecrdb, stoichiometry, microspecies, compounds, pkas, pkmgs, stan_codes
)
pkas.to_csv(os.path.join(here, OUTPUT_PATH_PKAS))
compounds.to_csv(os.path.join(here, OUTPUT_PATH_COMPOUNDS))
tecrdb.to_csv(os.path.join(here, OUTPUT_PATH_TECRDB))
jsondump(os.path.join(here, OUTPUT_PATH_STAN_CODES), stan_codes)
jsondump(os.path.join(here, OUTPUT_PATH_JSON), stan_input)
rdump(os.path.join(here, OUTPUT_PATH_R), stan_input)
def generate_samples(
study_name: str,
measurements: pd.DataFrame,
model_configurations: List[ModelConfiguration],
) -> None:
"""Run cmdstanpy.CmdStanModel.sample, do diagnostics and save results.
:param study_name: a string
"""
infds = {}
for model_config in model_configurations:
fit_name = f"{study_name}-{model_config.name}"
print(f"Fitting model {fit_name}...")
loo_file = os.path.join(LOO_DIR, f"loo_{fit_name}.pkl")
infd_file = os.path.join(INFD_DIR, f"infd_{fit_name}.ncdf")
json_file = os.path.join(JSON_DIR, f"input_data_{fit_name}.json")
stan_input = model_config.stan_input_function(measurements)
print(f"Writing input data to {json_file}")
jsondump(json_file, stan_input)
model = CmdStanModel(
model_name=fit_name, stan_file=model_config.stan_file
)
print(f"Writing csv files to {SAMPLES_DIR}...")
mcmc = model.sample(
data=stan_input,
output_dir=SAMPLES_DIR,
**model_config.sample_kwargs,
)
print(mcmc.diagnose().replace("\n\n", "\n"))
infd = az.from_cmdstanpy(
mcmc, **model_config.infd_kwargs_function(measurements)
)
print(az.summary(infd))
infds[fit_name] = infd
print(f"Writing inference data to {infd_file}")
infd.to_netcdf(infd_file)
print(f"Writing psis-loo results to {loo_file}\n")
az.loo(infd, pointwise=True).to_pickle(loo_file)
if len(infds) > 1:
comparison = az.compare(infds)
print(f"Loo comparison:\n{comparison}")
comparison.to_csv(os.path.join(LOO_DIR, "loo_comparison.csv"))
def test_jsondump(self):
def cmp(d1, d2):
self.assertEqual(d1.keys(), d2.keys())
for k in d1:
data_1 = d1[k]
data_2 = d2[k]
if isinstance(data_2, np.ndarray):
data_2 = data_2.tolist()
self.assertEqual(data_1, data_2)
dict_list = {'a': [1.0, 2.0, 3.0]}
file_list = os.path.join(_TMPDIR, 'list.json')
jsondump(file_list, dict_list)
with open(file_list) as fd:
cmp(json.load(fd), dict_list)
dict_vec = {'a': np.repeat(3, 4)}
file_vec = os.path.join(_TMPDIR, 'vec.json')
jsondump(file_vec, dict_vec)
with open(file_vec) as fd:
cmp(json.load(fd), dict_vec)
dict_zero_vec = {'a': []}
file_zero_vec = os.path.join(_TMPDIR, 'empty_vec.json')
jsondump(file_zero_vec, dict_zero_vec)
with open(file_zero_vec) as fd:
cmp(json.load(fd), dict_zero_vec)
dict_zero_matrix = {'a': [[], [], []]}
file_zero_matrix = os.path.join(_TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
cmp(json.load(fd), dict_zero_matrix)
arr = np.zeros(shape=(5, 0))
dict_zero_matrix = {'a': arr}
file_zero_matrix = os.path.join(_TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
cmp(json.load(fd), dict_zero_matrix)
def write_json(self, dict: Dict) -> None:
jsondump(self._data_file, dict)
def test_jsondump(self):
def cmp(d1, d2):
self.assertEqual(d1.keys(), d2.keys())
for k in d1:
data_1 = d1[k]
data_2 = d2[k]
if isinstance(data_2, np.ndarray):
data_2 = data_2.tolist()
self.assertEqual(data_1, data_2)
dict_list = {'a': [1.0, 2.0, 3.0]}
file_list = os.path.join(_TMPDIR, 'list.json')
jsondump(file_list, dict_list)
with open(file_list) as fd:
cmp(json.load(fd), dict_list)
dict_vec = {'a': np.repeat(3, 4)}
file_vec = os.path.join(_TMPDIR, 'vec.json')
jsondump(file_vec, dict_vec)
with open(file_vec) as fd:
cmp(json.load(fd), dict_vec)
dict_zero_vec = {'a': []}
file_zero_vec = os.path.join(_TMPDIR, 'empty_vec.json')
jsondump(file_zero_vec, dict_zero_vec)
with open(file_zero_vec) as fd:
cmp(json.load(fd), dict_zero_vec)
dict_zero_matrix = {'a': [[], [], []]}
file_zero_matrix = os.path.join(_TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
cmp(json.load(fd), dict_zero_matrix)
arr = np.zeros(shape=(5, 0))
dict_zero_matrix = {'a': arr}
file_zero_matrix = os.path.join(_TMPDIR, 'empty_matrix.json')
jsondump(file_zero_matrix, dict_zero_matrix)
with open(file_zero_matrix) as fd:
cmp(json.load(fd), dict_zero_matrix)
arr = np.zeros(shape=(2, 3, 4))
self.assertTrue(isinstance(arr, np.ndarray))
self.assertEqual(arr.shape, (2, 3, 4))
dict_3d_matrix = {'a': arr}
file_3d_matrix = os.path.join(_TMPDIR, '3d_matrix.json')
jsondump(file_3d_matrix, dict_3d_matrix)
with open(file_3d_matrix) as fd:
cmp(json.load(fd), dict_3d_matrix)
scalr = np.int32(1)
self.assertTrue(type(scalr).__module__ == 'numpy')
dict_scalr = {'a': scalr}
file_scalr = os.path.join(_TMPDIR, 'scalr.json')
jsondump(file_scalr, dict_scalr)
with open(file_scalr) as fd:
cmp(json.load(fd), dict_scalr)