Added the Fit benchmarking class (#1212)

* fixed the table docs

* updated the fitbenchmarking API in main

* added the Fit Class

* updated pylint config

* updated the tests for minimizers

* Updated the tests for main

* temp commit

* added return values to methods in fit class

* added more tests for the Fit class

* updated formating

* Updated tests

* deleted tests

* Updated docs

* updated tests

* removed the old functions

* Updated the patching for minimizer tests

* updating patching for hessian tests

* renamed file

* updated perform_fit method

* added more tests for benchmarking method to get full coverage

* refactored benchmarking and starting value tests

* refactored tests

* deleted starting value test file

* Updated the Software tests

* Updated loop_over_minimizers tests

* updated the jacobian tests

* added tests for the loop over hessian class

* renamed test file

* updated gitignore file

* added some more tests for the __perform_fit method

* Apply suggestions from code review

Co-authored-by: Letizia97 <letizia.protopapa@stfc.ac.uk>

* Added more tests

Co-authored-by: Letizia97 <letizia.protopapa@stfc.ac.uk>

* modified tests to work with os.path.join

* Updated files with linting fixes

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

* temp commit

---------

Co-authored-by: Letizia97 <letizia.protopapa@stfc.ac.uk>

.gitignore

@@ -14,3 +14,4 @@ docs/source/contributors/module_index
 .vscode/
 .eggs/
 *.out
+powermetrics_log.txt

fitbenchmarking/cli/main.py

@@ -15,7 +15,7 @@
 import fitbenchmarking
 from fitbenchmarking.cli.checkpoint_handler import generate_report
 from fitbenchmarking.cli.exception_handler import exception_handler
-from fitbenchmarking.core.fitting_benchmarking import benchmark
+from fitbenchmarking.core.fitting_benchmarking import Fit
 from fitbenchmarking.core.results_output import (create_index_page,
                                                  open_browser, save_results)
 from fitbenchmarking.utils.checkpoint import Checkpoint
@@ -293,11 +293,11 @@ def run(problem_sets, additional_options=None, options_file='', debug=False):
 
         LOGGER.info('Running the benchmarking on the %s problem set',
                     label)
-        results, failed_problems, unselected_minimizers = \
-            benchmark(options=options,
-                      data_dir=data_dir,
-                      label=label,
-                      checkpointer=cp)
+        fit = Fit(options=options,
+                  data_dir=data_dir,
+                  label=label,
+                  checkpointer=cp)
+        results, failed_problems, unselected_minimizers = fit.benchmark()
 
         # If a result has error flag 4 then the result contains dummy values,
         # if this is the case for all results then output should not be

fitbenchmarking/cli/tests/test_main.py

@@ -57,7 +57,7 @@ class TestMain(TestCase):
     Tests for main.py
     """
 
-    @patch('fitbenchmarking.cli.main.benchmark')
+    @patch('fitbenchmarking.cli.main.Fit.benchmark')
     def test_check_no_results_produced(self, benchmark):
         """
         Checks that exception is raised if no results are produced
@@ -68,7 +68,7 @@ def test_check_no_results_produced(self, benchmark):
             main.run(['examples/benchmark_problems/simple_tests'],
                      os.path.dirname(__file__), debug=True)
 
-    @patch('fitbenchmarking.cli.main.benchmark')
+    @patch('fitbenchmarking.cli.main.Fit.benchmark')
     def test_all_dummy_results_produced(self, benchmark):
         """
         Checks that exception is raised if all dummy results