Merge branch 'develop' into 2.37-tests

Author: WardBrian

.github/codecov.yml

@@ -0,0 +1 @@
+comment: false

.github/workflows/main.yml

@@ -49,7 +49,7 @@ jobs:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
     steps:
       - name: Check out github
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v5

.github/workflows/release.yml

@@ -18,7 +18,7 @@ jobs:
       fail-fast: false
     steps:
       - name: Check out source code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v5

cmdstanpy/__init__.py

@@ -42,6 +42,8 @@ def _cleanup_tmpdir() -> None:
     set_make_env,
     show_versions,
     write_stan_json,
+    enable_logging,
+    disable_logging,
 )
 
 __all__ = [
@@ -63,4 +65,6 @@ def _cleanup_tmpdir() -> None:
     'show_versions',
     'rebuild_cmdstan',
     'cmdstan_version',
+    "enable_logging",
+    "disable_logging",
 ]

cmdstanpy/stanfit/__init__.py

@@ -2,7 +2,7 @@
 
 import glob
 import os
-from typing import Any, Dict, List, Optional, Union
+from typing import List, Optional, Union
 
 from cmdstanpy.cmdstan_args import (
     CmdStanArgs,
@@ -12,7 +12,7 @@
     SamplerArgs,
     VariationalArgs,
 )
-from cmdstanpy.utils import check_sampler_csv, get_logger, scan_config
+from cmdstanpy.utils import check_sampler_csv, get_logger, stancsv
 
 from .gq import CmdStanGQ
 from .laplace import CmdStanLaplace
@@ -103,10 +103,9 @@ def from_csv(
                 ' includes non-csv file: {}'.format(file)
             )
 
-    config_dict: Dict[str, Any] = {}
     try:
-        with open(csvfiles[0], 'r') as fd:
-            scan_config(fd, config_dict, 0)
+        comments, *_ = stancsv.parse_comments_header_and_draws(csvfiles[0])
+        config_dict = stancsv.parse_config(comments)
     except (IOError, OSError, PermissionError) as e:
         raise ValueError('Cannot read CSV file: {}'.format(csvfiles[0])) from e
     if 'model' not in config_dict or 'method' not in config_dict:
@@ -118,39 +117,43 @@ def from_csv(
                 method, config_dict['method']
             )
         )
+    model: str = config_dict['model']  # type: ignore
     try:
         if config_dict['method'] == 'sample':
+            save_warmup = config_dict['save_warmup'] == 1
             chains = len(csvfiles)
+            num_samples: int = config_dict['num_samples']  # type: ignore
+            num_warmup: int = config_dict['num_warmup']  # type: ignore
+            thin: int = config_dict['thin']  # type: ignore
             sampler_args = SamplerArgs(
-                iter_sampling=config_dict['num_samples'],
-                iter_warmup=config_dict['num_warmup'],
-                thin=config_dict['thin'],
-                save_warmup=config_dict['save_warmup'],
+                iter_sampling=num_samples,
+                iter_warmup=num_warmup,
+                thin=thin,
+                save_warmup=save_warmup,
             )
             # bugfix 425, check for fixed_params output
             try:
                 check_sampler_csv(
                     csvfiles[0],
-                    iter_sampling=config_dict['num_samples'],
-                    iter_warmup=config_dict['num_warmup'],
-                    thin=config_dict['thin'],
-                    save_warmup=config_dict['save_warmup'],
+                    iter_sampling=num_samples,
+                    iter_warmup=num_warmup,
+                    thin=thin,
+                    save_warmup=save_warmup,
                 )
             except ValueError:
                 try:
                     check_sampler_csv(
                         csvfiles[0],
-                        is_fixed_param=True,
-                        iter_sampling=config_dict['num_samples'],
-                        iter_warmup=config_dict['num_warmup'],
-                        thin=config_dict['thin'],
-                        save_warmup=config_dict['save_warmup'],
+                        iter_sampling=num_samples,
+                        iter_warmup=num_warmup,
+                        thin=thin,
+                        save_warmup=save_warmup,
                     )
                     sampler_args = SamplerArgs(
-                        iter_sampling=config_dict['num_samples'],
-                        iter_warmup=config_dict['num_warmup'],
-                        thin=config_dict['thin'],
-                        save_warmup=config_dict['save_warmup'],
+                        iter_sampling=num_samples,
+                        iter_warmup=num_warmup,
+                        thin=thin,
+                        save_warmup=save_warmup,
                         fixed_param=True,
                     )
                 except ValueError as e:
@@ -159,8 +162,8 @@ def from_csv(
                     ) from e
 
             cmdstan_args = CmdStanArgs(
-                model_name=config_dict['model'],
-                model_exe=config_dict['model'],
+                model_name=model,
+                model_exe=model,
                 chain_ids=[x + 1 for x in range(chains)],
                 method_args=sampler_args,
             )
@@ -177,14 +180,18 @@ def from_csv(
                     "Cannot find optimization algorithm"
                     " in file {}.".format(csvfiles[0])
                 )
+            algorithm: str = config_dict['algorithm']  # type: ignore
+            save_iterations = config_dict['save_iterations'] == 1
+            jacobian = config_dict.get('jacobian', 0) == 1
+
             optimize_args = OptimizeArgs(
-                algorithm=config_dict['algorithm'],
-                save_iterations=config_dict['save_iterations'],
-                jacobian=config_dict.get('jacobian', 0),
+                algorithm=algorithm,
+                save_iterations=save_iterations,
+                jacobian=jacobian,
             )
             cmdstan_args = CmdStanArgs(
-                model_name=config_dict['model'],
-                model_exe=config_dict['model'],
+                model_name=model,
+                model_exe=model,
                 chain_ids=None,
                 method_args=optimize_args,
             )
@@ -200,18 +207,18 @@ def from_csv(
                     " in file {}.".format(csvfiles[0])
                 )
             variational_args = VariationalArgs(
-                algorithm=config_dict['algorithm'],
-                iter=config_dict['iter'],
-                grad_samples=config_dict['grad_samples'],
-                elbo_samples=config_dict['elbo_samples'],
-                eta=config_dict['eta'],
-                tol_rel_obj=config_dict['tol_rel_obj'],
-                eval_elbo=config_dict['eval_elbo'],
-                output_samples=config_dict['output_samples'],
+                algorithm=config_dict['algorithm'],  # type: ignore
+                iter=config_dict['iter'],  # type: ignore
+                grad_samples=config_dict['grad_samples'],  # type: ignore
+                elbo_samples=config_dict['elbo_samples'],  # type: ignore
+                eta=config_dict['eta'],  # type: ignore
+                tol_rel_obj=config_dict['tol_rel_obj'],  # type: ignore
+                eval_elbo=config_dict['eval_elbo'],  # type: ignore
+                output_samples=config_dict['output_samples'],  # type: ignore
             )
             cmdstan_args = CmdStanArgs(
-                model_name=config_dict['model'],
-                model_exe=config_dict['model'],
+                model_name=model,
+                model_exe=model,
                 chain_ids=None,
                 method_args=variational_args,
             )
@@ -221,14 +228,15 @@ def from_csv(
                 runset._set_retcode(i, 0)
             return CmdStanVB(runset)
         elif config_dict['method'] == 'laplace':
+            jacobian = config_dict['jacobian'] == 1
             laplace_args = LaplaceArgs(
-                mode=config_dict['mode'],
-                draws=config_dict['draws'],
-                jacobian=config_dict['jacobian'],
+                mode=config_dict['mode'],  # type: ignore
+                draws=config_dict['draws'],  # type: ignore
+                jacobian=jacobian,
             )
             cmdstan_args = CmdStanArgs(
-                model_name=config_dict['model'],
-                model_exe=config_dict['model'],
+                model_name=model,
+                model_exe=model,
                 chain_ids=None,
                 method_args=laplace_args,
             )
@@ -237,18 +245,18 @@ def from_csv(
             for i in range(len(runset._retcodes)):
                 runset._set_retcode(i, 0)
             mode: CmdStanMLE = from_csv(
-                config_dict['mode'],
+                config_dict['mode'],  # type: ignore
                 method='optimize',
             )  # type: ignore
             return CmdStanLaplace(runset, mode=mode)
         elif config_dict['method'] == 'pathfinder':
             pathfinder_args = PathfinderArgs(
-                num_draws=config_dict['num_draws'],
-                num_paths=config_dict['num_paths'],
+                num_draws=config_dict['num_draws'],  # type: ignore
+                num_paths=config_dict['num_paths'],  # type: ignore
             )
             cmdstan_args = CmdStanArgs(
-                model_name=config_dict['model'],
-                model_exe=config_dict['model'],
+                model_name=model,
+                model_exe=model,
                 chain_ids=None,
                 method_args=pathfinder_args,
             )

cmdstanpy/stanfit/gq.py

@@ -31,8 +31,12 @@
 
 
 from cmdstanpy.cmdstan_args import Method
-from cmdstanpy.utils import build_xarray_data, flatten_chains, get_logger
-from cmdstanpy.utils.stancsv import scan_generic_csv
+from cmdstanpy.utils import (
+    build_xarray_data,
+    flatten_chains,
+    get_logger,
+    stancsv,
+)
 
 from .mcmc import CmdStanMCMC
 from .metadata import InferenceMetadata
@@ -65,8 +69,7 @@ def __init__(
         self.previous_fit: Fit = previous_fit
 
         self._draws: np.ndarray = np.array(())
-        config = self._validate_csv_files()
-        self._metadata = InferenceMetadata(config)
+        self._metadata = self._validate_csv_files()
 
     def __repr__(self) -> str:
         repr = 'CmdStanGQ: model={} chains={}{}'.format(
@@ -99,48 +102,38 @@ def __getstate__(self) -> dict:
         self._assemble_generated_quantities()
         return self.__dict__
 
-    def _validate_csv_files(self) -> Dict[str, Any]:
+    def _validate_csv_files(self) -> InferenceMetadata:
         """
         Checks that Stan CSV output files for all chains are consistent
-        and returns dict containing config and column names.
+        and returns InferenceMetadata object containing config and column names.
 
-        Raises exception when inconsistencies detected.
+        Raises exception if inconsistencies are detected.
         """
-        dzero = {}
-        for i in range(self.chains):
-            if i == 0:
-                dzero = scan_generic_csv(
-                    path=self.runset.csv_files[i],
-                )
-            else:
-                drest = scan_generic_csv(
-                    path=self.runset.csv_files[i],
-                )
-                for key in dzero:
-                    if (
-                        key
-                        not in [
-                            'id',
-                            'fitted_params',
-                            'diagnostic_file',
-                            'metric_file',
-                            'profile_file',
-                            'init',
-                            'seed',
-                            'start_datetime',
-                        ]
-                        and dzero[key] != drest[key]
-                    ):
-                        raise ValueError(
-                            'CmdStan config mismatch in Stan CSV file {}: '
-                            'arg {} is {}, expected {}'.format(
-                                self.runset.csv_files[i],
-                                key,
-                                dzero[key],
-                                drest[key],
-                            )
+        excluded_fields = {
+            'id',
+            'fitted_params',
+            'diagnostic_file',
+            'metric_file',
+            'profile_file',
+            'init',
+            'seed',
+            'start_datetime',
+        }
+        meta0 = InferenceMetadata.from_csv(self.runset.csv_files[0])
+        for i in range(1, self.chains):
+            meta = InferenceMetadata.from_csv(self.runset.csv_files[i])
+            for key in set(meta._cmdstan_config.keys()) - excluded_fields:
+                if meta0[key] != meta[key]:
+                    raise ValueError(
+                        'CmdStan config mismatch in Stan CSV file {}: '
+                        'arg {} is {}, expected {}'.format(
+                            self.runset.csv_files[i],
+                            key,
+                            meta0[key],
+                            meta[key],
                         )
-        return dzero
+                    )
+        return meta0
 
     @property
     def chains(self) -> int:
@@ -157,7 +150,7 @@ def column_names(self) -> Tuple[str, ...]:
         """
         Names of generated quantities of interest.
         """
-        return self._metadata.cmdstan_config['column_names']  # type: ignore
+        return self._metadata.column_names
 
     @property
     def metadata(self) -> InferenceMetadata:
@@ -633,11 +626,17 @@ def _assemble_generated_quantities(self) -> None:
             order='F',
         )
         for chain in range(self.chains):
-            with open(self.runset.csv_files[chain], 'r') as fd:
-                lines = (line for line in fd if not line.startswith('#'))
-                gq_sample[:, chain, :] = np.loadtxt(
-                    lines, dtype=np.ndarray, ndmin=2, skiprows=1, delimiter=','
+            csv_file = self.runset.csv_files[chain]
+            try:
+                *_, draws = stancsv.parse_comments_header_and_draws(
+                    self.runset.csv_files[chain]
                 )
+                gq_sample[:, chain, :] = stancsv.csv_bytes_list_to_numpy(draws)
+            except Exception as exc:
+                raise ValueError(
+                    f"An error occurred when parsing Stan csv {csv_file}"
+                    f" for chain {chain}"
+                ) from exc
         self._draws = gq_sample
 
     def _draws_start(self, inc_warmup: bool) -> Tuple[int, int]: