\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-bcfu571m because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-kmcy1dh2 because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n
\n
\n
\n
[2]:\n
\n
\n
print('a =',fit_result.value(a))\n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -33,17 +33,17 @@\n }\n model = CallableNumericalModel(model_dict, connectivity_mapping={z: {y, b}, y:\n {x, a}})\n \n # Apply model\n fit = Fit(model, x=x_data, z=z_data)\n fit_result = fit.execute()\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-bcfu571m\n-because the default path (/nonexistent/first-build/.config/matplotlib) is not a\n-writable directory; it is highly recommended to set the MPLCONFIGDIR\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-kmcy1dh2\n+because the default path (/nonexistent/second-build/.config/matplotlib) is not\n+a writable directory; it is highly recommended to set the MPLCONFIGDIR\n environment variable to a writable directory, in particular to speed up the\n import of Matplotlib and to better support multiprocessing.\n [2]:\n print('a =', fit_result.value(a))\n a = 0.599999973796809\n [3]:\n print('b =', fit_result.value(b))\n"}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_CallableNumericalModel_ode.ipynb.gz", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_CallableNumericalModel_ode.ipynb.gz", "unified_diff": null, "details": [{"source1": "ex_CallableNumericalModel_ode.ipynb", "source2": "ex_CallableNumericalModel_ode.ipynb", "unified_diff": null, "details": [{"source1": "Pretty-printed", "source2": "Pretty-printed", "comments": ["Similarity: 0.9971354166666666%", "Differences: {\"'cells'\": \"{1: {'metadata': {'execution': {'iopub.execute_input': '2024-09-03T04:38:15.198362Z', \"", " \"'iopub.status.busy': '2024-09-03T04:38:15.198041Z', 'iopub.status.idle': \"", " \"'2024-09-03T04:38:19.047103Z', 'shell.execute_reply': \"", " \"'2024-09-03T04:38:19.046101Z'}}, 'outputs': {0: {'text': ['Matplotlib created a \"", " 'temporary cache directory at /tmp/matplotlib-kmcy1dh2 because the default path '", " '(/nonexistent/second-build/.config/matplotlib) is no [\u2026]"], "unified_diff": "@@ -16,26 +16,26 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:45.806578Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:45.806126Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:48.154615Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:48.153910Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:15.198362Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:15.198041Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:19.047103Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:19.046101Z\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stderr\",\n \"output_type\": \"stream\",\n \"text\": [\n- \"Matplotlib created a temporary cache directory at /tmp/matplotlib-bcfu571m because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n+ \"Matplotlib created a temporary cache directory at /tmp/matplotlib-kmcy1dh2 because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n ]\n }\n ],\n \"source\": [\n \"from symfit import variables, Parameter, Fit, D, ODEModel, CallableNumericalModel\\n\",\n \"import numpy as np\\n\",\n \"import matplotlib.pyplot as plt\\n\",\n@@ -65,18 +65,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 2,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:48.158080Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:48.157558Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:48.161284Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:48.160741Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:19.053264Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:19.052799Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:19.058077Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:19.057157Z\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stdout\",\n \"output_type\": \"stream\",\n \"text\": [\n@@ -89,18 +89,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 3,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:48.163763Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:48.163284Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:48.166448Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:48.165940Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:19.061447Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:19.061121Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:19.065755Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:19.064876Z\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stdout\",\n \"output_type\": \"stream\",\n \"text\": [\n"}]}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_bivariate_likelihood.html", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_bivariate_likelihood.html", "unified_diff": "@@ -53,15 +53,15 @@\n
\n
\n
\n
\n
\n
\n
\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-6tc_pvm1 because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-1jn2rwew because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n
\n
\n
Build a model corresponding to a bivariate normal distribution.
\n
\n
[2]:\n
\n
\n@@ -146,16 +146,16 @@\n rho 6.026420e-01 2.013810e-03\n sig_x 1.100898e-01 2.461684e-04\n sig_y 2.303400e-01 5.150556e-04\n x0 5.901317e-01 3.481346e-04\n y0 8.014040e-01 7.283990e-04\n Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\n Number of iterations 22\n-Objective <symfit.core.objectives.LogLikelihood object at 0xf0bd2630>\n-Minimizer <symfit.core.minimizers.LBFGSB object at 0xeaf56e70>\n+Objective <symfit.core.objectives.LogLikelihood object at 0xf0b4b570>\n+Minimizer <symfit.core.minimizers.LBFGSB object at 0xe9c38990>\n \n Goodness of fit qualifiers:\n likelihood inf\n log_likelihood 106241.24669486462\n objective_value -106241.24669486462\n
\n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -8,17 +8,17 @@\n [1]:\n import numpy as np\n from symfit import Variable, Parameter, Fit\n from symfit.core.objectives import LogLikelihood\n from symfit.distributions import BivariateGaussian\n import matplotlib.pyplot as plt\n \n-Matplotlib created a temporary cache directory at /tmp/matplotlib-6tc_pvm1\n-because the default path (/nonexistent/first-build/.config/matplotlib) is not a\n-writable directory; it is highly recommended to set the MPLCONFIGDIR\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-1jn2rwew\n+because the default path (/nonexistent/second-build/.config/matplotlib) is not\n+a writable directory; it is highly recommended to set the MPLCONFIGDIR\n environment variable to a writable directory, in particular to speed up the\n import of Matplotlib and to better support multiprocessing.\n Build a model corresponding to a bivariate normal distribution.\n [2]:\n x = Variable('x')\n y = Variable('y')\n x0 = Parameter('x0', value=0.6, min=0.5, max=0.7)\n@@ -71,16 +71,16 @@\n sig_x 1.100898e-01 2.461684e-04\n sig_y 2.303400e-01 5.150556e-04\n x0 5.901317e-01 3.481346e-04\n y0 8.014040e-01 7.283990e-04\n Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\n Number of iterations 22\n Objective \n-Minimizer \n+0xf0b4b570>\n+Minimizer \n \n Goodness of fit qualifiers:\n likelihood inf\n log_likelihood 106241.24669486462\n objective_value -106241.24669486462\n We see that this result is in agreement with our data.\n *\b**\b**\b**\b**\b**\b* _\bs\bs_\by\by_\bm\bm_\bf\bf_\bi\bi_\bt\bt *\b**\b**\b**\b**\b**\b*\n"}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_bivariate_likelihood.ipynb.gz", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_bivariate_likelihood.ipynb.gz", "unified_diff": null, "details": [{"source1": "ex_bivariate_likelihood.ipynb", "source2": "ex_bivariate_likelihood.ipynb", "unified_diff": null, "details": [{"source1": "Pretty-printed", "source2": "Pretty-printed", "comments": ["Similarity: 0.9989949845679013%", "Differences: {\"'cells'\": \"{1: {'metadata': {'execution': {'iopub.execute_input': '2024-09-03T04:38:21.825590Z', \"", " \"'iopub.status.busy': '2024-09-03T04:38:21.825240Z', 'iopub.status.idle': \"", " \"'2024-09-03T04:38:23.640215Z', 'shell.execute_reply': \"", " \"'2024-09-03T04:38:23.639197Z'}}, 'outputs': {0: {'text': ['Matplotlib created a \"", " 'temporary cache directory at /tmp/matplotlib-1jn2rwew because the default path '", " '(/nonexistent/second-build/.config/matplotlib) is no [\u2026]"], "unified_diff": "@@ -22,31 +22,31 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:49.899770Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:49.899284Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:51.037664Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:51.037031Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:21.825590Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:21.825240Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:23.640215Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:23.639197Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stderr\",\n \"output_type\": \"stream\",\n \"text\": [\n- \"Matplotlib created a temporary cache directory at /tmp/matplotlib-6tc_pvm1 because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n+ \"Matplotlib created a temporary cache directory at /tmp/matplotlib-1jn2rwew because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n ]\n }\n ],\n \"source\": [\n \"import numpy as np\\n\",\n \"from symfit import Variable, Parameter, Fit\\n\",\n \"from symfit.core.objectives import LogLikelihood\\n\",\n@@ -67,18 +67,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 2,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:51.041112Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:51.040592Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:51.104276Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:51.103685Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:23.644906Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:23.644235Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:23.751171Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:23.750195Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -109,18 +109,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 3,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:51.107425Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:51.107006Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:51.127973Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:51.127419Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:23.755680Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:23.755362Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:23.794118Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:23.793093Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -136,18 +136,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 4,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:51.130725Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:51.130266Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:53:51.317430Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:53:51.316855Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:23.797561Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:23.797234Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:24.089734Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:24.088782Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -180,18 +180,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 5,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:53:51.320283Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:53:51.319839Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:01.213668Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:01.213021Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:24.093289Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:24.092979Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:39.128876Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:39.128116Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -227,16 +227,16 @@\n \"rho 6.026420e-01 2.013810e-03\\n\",\n \"sig_x 1.100898e-01 2.461684e-04\\n\",\n \"sig_y 2.303400e-01 5.150556e-04\\n\",\n \"x0 5.901317e-01 3.481346e-04\\n\",\n \"y0 8.014040e-01 7.283990e-04\\n\",\n \"Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\\n\",\n \"Number of iterations 22\\n\",\n- \"Objective \\n\",\n- \"Minimizer \\n\",\n+ \"Objective \\n\",\n+ \"Minimizer \\n\",\n \"\\n\",\n \"Goodness of fit qualifiers:\\n\",\n \"likelihood inf\\n\",\n \"log_likelihood 106241.24669486462\\n\",\n \"objective_value -106241.24669486462\\n\"\n ]\n }\n"}]}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_mexican_hat.html", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_mexican_hat.html", "unified_diff": "@@ -51,15 +51,15 @@\n \n \n
\n
\n
\n
\n
\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-7den1jfx because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-duj97rbb because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n
\n
\n
First we define a model for the skewed mexican hat.
\n
\n
[2]:\n
\n
\n@@ -99,15 +99,15 @@\n
\n
\n
[3]:\n
\n
\n
\n
\n-<matplotlib.legend.Legend at 0xe9eaaaf8>\n+<matplotlib.legend.Legend at 0xe9bbdb10>\n
\n
\n
\n
\n
\n
\n \n@@ -169,15 +169,15 @@\n
\n
\n
\n
\n
\n
\n exact value -2.35191046133532\n-num value -2.356519816345737\n+num value -2.3555865214239358\n
\n
\n
Using DifferentialEvolution, we find the correct global minimum. However, it is not exactly the same as the analytical solution. This is because DifferentialEvolution is expensive to perform, and therefore does not solve to high precision by default. We could demand a higher precission from DifferentialEvolution, but this isn\u2019t worth the high computational cost. Instead, we will just tell symfit to perform DifferentialEvolution, followed by BFGS.
\n
\n
[7]:\n
\n
\n@@ -190,15 +190,15 @@\n
\n
\n
\n
\n
\n
\n exact value -2.35191046133532\n-num value -2.351910461335324\n+num value -2.351910461335322\n
\n
\n
We see that now the proper solution has been found to much higher precision.
\n \n \n \n
\n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,17 +5,17 @@\n then use DifferentialEvolution to find the global minimum.\n [1]:\n from symfit import Parameter, Variable, Model, Fit, solve, diff, N, re\n from symfit.core.minimizers import DifferentialEvolution, BFGS\n import numpy as np\n import matplotlib.pyplot as plt\n \n-Matplotlib created a temporary cache directory at /tmp/matplotlib-7den1jfx\n-because the default path (/nonexistent/first-build/.config/matplotlib) is not a\n-writable directory; it is highly recommended to set the MPLCONFIGDIR\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-duj97rbb\n+because the default path (/nonexistent/second-build/.config/matplotlib) is not\n+a writable directory; it is highly recommended to set the MPLCONFIGDIR\n environment variable to a writable directory, in particular to speed up the\n import of Matplotlib and to better support multiprocessing.\n First we define a model for the skewed mexican hat.\n [2]:\n x = Parameter('x')\n x.min, x.max = -100, 100\n y = Variable('y')\n@@ -33,15 +33,15 @@\n plt.plot(xdata, ydata, label=r'$f(x)$')\n plt.xlabel('x')\n plt.ylabel('f(x)')\n plt.ylim(1.1 * ydata.min(), 1.1 * ydata.max())\n plt.legend()\n \n [3]:\n-\n+\n [../_images/examples_ex_mexican_hat_5_1.png]\n Using sympy, it is easy to solve the solution analytically, by finding the\n places where the gradient is zero.\n [4]:\n sol = solve(diff(model[y], x), x)\n # Give numerical value\n sol = [re(N(s)) for s in sol]\n@@ -61,29 +61,29 @@\n [6]:\n fit = Fit(model, minimizer=DifferentialEvolution)\n fit_result = fit.execute()\n print('exact value', sol[2])\n print('num value ', fit_result.value(x))\n \n exact value -2.35191046133532\n-num value -2.356519816345737\n+num value -2.3555865214239358\n Using DifferentialEvolution, we find the correct global minimum. However, it is\n not exactly the same as the analytical solution. This is because\n DifferentialEvolution is expensive to perform, and therefore does not solve to\n high precision by default. We could demand a higher precission from\n DifferentialEvolution, but this isn\u2019t worth the high computational cost.\n Instead, we will just tell symfit to perform DifferentialEvolution, followed by\n BFGS.\n [7]:\n fit = Fit(model, minimizer=[DifferentialEvolution, BFGS])\n fit_result = fit.execute()\n print('exact value', sol[2])\n print('num value ', fit_result.value(x))\n exact value -2.35191046133532\n-num value -2.351910461335324\n+num value -2.351910461335322\n We see that now the proper solution has been found to much higher precision.\n *\b**\b**\b**\b**\b**\b* _\bs\bs_\by\by_\bm\bm_\bf\bf_\bi\bi_\bt\bt *\b**\b**\b**\b**\b**\b*\n *\b**\b**\b**\b* N\bNa\bav\bvi\big\bga\bat\bti\bio\bon\bn *\b**\b**\b**\b*\n * _\bI_\bn_\bt_\br_\bo_\bd_\bu_\bc_\bt_\bi_\bo_\bn\n * _\bI_\bn_\bs_\bt_\ba_\bl_\bl_\ba_\bt_\bi_\bo_\bn\n * _\bT_\bu_\bt_\bo_\br_\bi_\ba_\bl\n * _\bF_\bi_\bt_\bt_\bi_\bn_\bg_\b _\bT_\by_\bp_\be_\bs\n"}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_mexican_hat.ipynb.gz", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_mexican_hat.ipynb.gz", "unified_diff": null, "details": [{"source1": "ex_mexican_hat.ipynb", "source2": "ex_mexican_hat.ipynb", "unified_diff": null, "details": [{"source1": "Pretty-printed", "source2": "Pretty-printed", "comments": ["Similarity: 0.9989207175925926%", "Differences: {\"'cells'\": \"{1: {'metadata': {'execution': {'iopub.execute_input': '2024-09-03T04:38:41.911974Z', \"", " \"'iopub.status.busy': '2024-09-03T04:38:41.911642Z', 'iopub.status.idle': \"", " \"'2024-09-03T04:38:43.765011Z', 'shell.execute_reply': \"", " \"'2024-09-03T04:38:43.764017Z'}}, 'outputs': {0: {'text': ['Matplotlib created a \"", " 'temporary cache directory at /tmp/matplotlib-duj97rbb because the default path '", " '(/nonexistent/second-build/.config/matplotlib) is no [\u2026]"], "unified_diff": "@@ -15,29 +15,29 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:02.996560Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:02.996049Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.173060Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.172433Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:41.911974Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:41.911642Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:43.765011Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:43.764017Z\"\n },\n \"pycharm\": {\n \"is_executing\": false\n }\n },\n \"outputs\": [\n {\n \"name\": \"stderr\",\n \"output_type\": \"stream\",\n \"text\": [\n- \"Matplotlib created a temporary cache directory at /tmp/matplotlib-7den1jfx because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n+ \"Matplotlib created a temporary cache directory at /tmp/matplotlib-duj97rbb because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n ]\n }\n ],\n \"source\": [\n \"from symfit import Parameter, Variable, Model, Fit, solve, diff, N, re\\n\",\n \"from symfit.core.minimizers import DifferentialEvolution, BFGS\\n\",\n \"import numpy as np\\n\",\n@@ -57,18 +57,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 2,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.176578Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.175838Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.212102Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.211541Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:43.769903Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:43.769159Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:43.829876Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:43.828942Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -101,30 +101,30 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 3,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.214938Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.214338Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.490215Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.489659Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:43.833629Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:43.833342Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:44.252922Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:44.252136Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"data\": {\n \"text/plain\": [\n- \"\"\n+ \"\"\n ]\n },\n \"execution_count\": 3,\n \"metadata\": {},\n \"output_type\": \"execute_result\"\n },\n {\n@@ -163,18 +163,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 4,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.492983Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.492524Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.610116Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.609562Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:44.256657Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:44.256332Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:44.444911Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:44.443937Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -210,18 +210,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 5,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.612776Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.612340Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.640316Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.639768Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:44.448651Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:44.448331Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:44.487474Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:44.486502Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -256,32 +256,32 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 6,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.643013Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.642457Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.763437Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.762862Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:44.491645Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:44.491330Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:44.657545Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:44.656425Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stdout\",\n \"output_type\": \"stream\",\n \"text\": [\n \"exact value -2.35191046133532\\n\",\n- \"num value -2.356519816345737\\n\"\n+ \"num value -2.3555865214239358\\n\"\n ]\n }\n ],\n \"source\": [\n \"fit = Fit(model, minimizer=DifferentialEvolution)\\n\",\n \"fit_result = fit.execute()\\n\",\n \"print('exact value', sol[2])\\n\",\n@@ -300,32 +300,32 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 7,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:04.766241Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:04.765701Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:04.905024Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:04.904456Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:44.661225Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:44.660916Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:44.849432Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:44.848487Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stdout\",\n \"output_type\": \"stream\",\n \"text\": [\n \"exact value -2.35191046133532\\n\",\n- \"num value -2.351910461335324\\n\"\n+ \"num value -2.351910461335322\\n\"\n ]\n }\n ],\n \"source\": [\n \"fit = Fit(model, minimizer=[DifferentialEvolution, BFGS])\\n\",\n \"fit_result = fit.execute()\\n\",\n \"print('exact value', sol[2])\\n\",\n"}]}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_ode_system.html", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_ode_system.html", "unified_diff": "@@ -53,15 +53,15 @@\n \n \n
\n
\n
\n
\n
\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-ctf0_1nl because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-ffgb4lfg because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n
\n
\n
First we build a model representing the system of equations.
\n
\n
[2]:\n
\n
\n@@ -146,27 +146,27 @@\n
\n
\n
\n
\n
\n \n Parameter Value Standard Deviation\n-k1_f 9.540415e-02 4.440702e-03\n-k1_r 1.065111e-01 7.165776e-02\n-k2_f 2.706139e-01 5.305094e-02\n-k2_r 2.633638e-01 5.647281e-02\n+k1_f 9.540412e-02 4.440666e-03\n+k1_r 1.065101e-01 7.165687e-02\n+k2_f 2.706136e-01 5.305083e-02\n+k2_r 2.633634e-01 5.647270e-02\n Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\n-Number of iterations 31\n-Objective <symfit.core.objectives.LeastSquares object at 0xe9c60ed0>\n-Minimizer <symfit.core.minimizers.LBFGSB object at 0xe9c60f60>\n+Number of iterations 30\n+Objective <symfit.core.objectives.LeastSquares object at 0xe998a4b0>\n+Minimizer <symfit.core.minimizers.LBFGSB object at 0xe998a570>\n \n Goodness of fit qualifiers:\n-chi_squared 33.98549453535547\n-objective_value 16.992747267677736\n-r_squared 0.9936568366399678\n+chi_squared 33.98549453602719\n+objective_value 16.992747268013595\n+r_squared 0.9936568366398425\n
\n
\n
\n
[5]:\n
\n
\n
taxis=np.linspace(tdata.min(),tdata.max(),1000)\n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -6,17 +6,17 @@\n from symfit import (\n variables, parameters, ODEModel, D, Fit\n )\n from symfit.core.support import key2str\n import numpy as np\n import matplotlib.pyplot as plt\n \n-Matplotlib created a temporary cache directory at /tmp/matplotlib-ctf0_1nl\n-because the default path (/nonexistent/first-build/.config/matplotlib) is not a\n-writable directory; it is highly recommended to set the MPLCONFIGDIR\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-ffgb4lfg\n+because the default path (/nonexistent/second-build/.config/matplotlib) is not\n+a writable directory; it is highly recommended to set the MPLCONFIGDIR\n environment variable to a writable directory, in particular to speed up the\n import of Matplotlib and to better support multiprocessing.\n First we build a model representing the system of equations.\n [2]:\n t, F, MM, FMM, FMMF = variables('t, F, MM, FMM, FMMF')\n k1_f, k1_r, k2_f, k2_r = parameters('k1_f, k1_r, k2_f, k2_r')\n \n@@ -70,28 +70,28 @@\n fit = Fit(model, t=tdata, MM=data[MM], F=data[F],\n FMMF=None, FMM=None,\n sigma_F=sigma_data, sigma_MM=sigma_data)\n fit_result = fit.execute()\n print(fit_result)\n \n Parameter Value Standard Deviation\n-k1_f 9.540415e-02 4.440702e-03\n-k1_r 1.065111e-01 7.165776e-02\n-k2_f 2.706139e-01 5.305094e-02\n-k2_r 2.633638e-01 5.647281e-02\n+k1_f 9.540412e-02 4.440666e-03\n+k1_r 1.065101e-01 7.165687e-02\n+k2_f 2.706136e-01 5.305083e-02\n+k2_r 2.633634e-01 5.647270e-02\n Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\n-Number of iterations 31\n+Number of iterations 30\n Objective \n-Minimizer \n+0xe998a4b0>\n+Minimizer \n \n Goodness of fit qualifiers:\n-chi_squared 33.98549453535547\n-objective_value 16.992747267677736\n-r_squared 0.9936568366399678\n+chi_squared 33.98549453602719\n+objective_value 16.992747268013595\n+r_squared 0.9936568366398425\n [5]:\n taxis = np.linspace(tdata.min(), tdata.max(), 1000)\n model_fit = model(t=taxis, **fit_result.params)._asdict()\n for var in data:\n plt.scatter(tdata, data[var], label='[{}]'.format(var.name))\n plt.plot(taxis, model_fit[var], label='[{}]'.format(var.name))\n plt.legend()\n"}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_ode_system.ipynb.gz", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_ode_system.ipynb.gz", "unified_diff": null, "details": [{"source1": "ex_ode_system.ipynb", "source2": "ex_ode_system.ipynb", "unified_diff": null, "details": [{"source1": "Pretty-printed", "source2": "Pretty-printed", "comments": ["Similarity: 0.9987916666666666%", "Differences: {\"'cells'\": \"{1: {'metadata': {'execution': {'iopub.execute_input': '2024-09-03T04:38:47.644225Z', \"", " \"'iopub.status.busy': '2024-09-03T04:38:47.643909Z', 'iopub.status.idle': \"", " \"'2024-09-03T04:38:49.432040Z', 'shell.execute_reply': \"", " \"'2024-09-03T04:38:49.431016Z'}}, 'outputs': {0: {'text': ['Matplotlib created a \"", " 'temporary cache directory at /tmp/matplotlib-ffgb4lfg because the default path '", " '(/nonexistent/second-build/.config/matplotlib) is no [\u2026]"], "unified_diff": "@@ -13,29 +13,29 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:06.637298Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:06.636897Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:07.977193Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:07.976565Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:47.644225Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:47.643909Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:49.432040Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:49.431016Z\"\n },\n \"pycharm\": {\n \"is_executing\": false\n }\n },\n \"outputs\": [\n {\n \"name\": \"stderr\",\n \"output_type\": \"stream\",\n \"text\": [\n- \"Matplotlib created a temporary cache directory at /tmp/matplotlib-ctf0_1nl because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n+ \"Matplotlib created a temporary cache directory at /tmp/matplotlib-ffgb4lfg because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\\n\"\n ]\n }\n ],\n \"source\": [\n \"from symfit import (\\n\",\n \"\\tvariables, parameters, ODEModel, D, Fit\\n\",\n \") \\n\",\n@@ -58,18 +58,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 2,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:07.980359Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:07.979845Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:08.023176Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:08.022591Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:49.436701Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:49.436036Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:49.509513Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:49.508476Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -118,18 +118,18 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 3,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:08.025864Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:08.025456Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:08.471367Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:08.470556Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:49.513271Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:49.512949Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:38:50.037729Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:38:50.036797Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n@@ -177,45 +177,45 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 4,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:08.474329Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:08.473870Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:16.505669Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:16.504927Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:38:50.041592Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:38:50.041230Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:39:02.761143Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:39:02.760030Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"name\": \"stdout\",\n \"output_type\": \"stream\",\n \"text\": [\n \"\\n\",\n \"Parameter Value Standard Deviation\\n\",\n- \"k1_f 9.540415e-02 4.440702e-03\\n\",\n- \"k1_r 1.065111e-01 7.165776e-02\\n\",\n- \"k2_f 2.706139e-01 5.305094e-02\\n\",\n- \"k2_r 2.633638e-01 5.647281e-02\\n\",\n+ \"k1_f 9.540412e-02 4.440666e-03\\n\",\n+ \"k1_r 1.065101e-01 7.165687e-02\\n\",\n+ \"k2_f 2.706136e-01 5.305083e-02\\n\",\n+ \"k2_r 2.633634e-01 5.647270e-02\\n\",\n \"Status message CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\\n\",\n- \"Number of iterations 31\\n\",\n- \"Objective \\n\",\n- \"Minimizer \\n\",\n+ \"Number of iterations 30\\n\",\n+ \"Objective \\n\",\n+ \"Minimizer \\n\",\n \"\\n\",\n \"Goodness of fit qualifiers:\\n\",\n- \"chi_squared 33.98549453535547\\n\",\n- \"objective_value 16.992747267677736\\n\",\n- \"r_squared 0.9936568366399678\\n\"\n+ \"chi_squared 33.98549453602719\\n\",\n+ \"objective_value 16.992747268013595\\n\",\n+ \"r_squared 0.9936568366398425\\n\"\n ]\n }\n ],\n \"source\": [\n \"k1_f.min, k1_f.max = 0, 1\\n\",\n \"k1_r.min, k1_r.max = 0, 1\\n\",\n \"k2_f.min, k2_f.max = 0, 1\\n\",\n@@ -229,29 +229,29 @@\n ]\n },\n {\n \"cell_type\": \"code\",\n \"execution_count\": 5,\n \"metadata\": {\n \"execution\": {\n- \"iopub.execute_input\": \"2025-10-06T10:54:16.508404Z\",\n- \"iopub.status.busy\": \"2025-10-06T10:54:16.508168Z\",\n- \"iopub.status.idle\": \"2025-10-06T10:54:16.800508Z\",\n- \"shell.execute_reply\": \"2025-10-06T10:54:16.799941Z\"\n+ \"iopub.execute_input\": \"2024-09-03T04:39:02.765160Z\",\n+ \"iopub.status.busy\": \"2024-09-03T04:39:02.764791Z\",\n+ \"iopub.status.idle\": \"2024-09-03T04:39:03.249650Z\",\n+ \"shell.execute_reply\": \"2024-09-03T04:39:03.248606Z\"\n },\n \"pycharm\": {\n \"is_executing\": false,\n \"metadata\": false,\n \"name\": \"#%%\\n\"\n }\n },\n \"outputs\": [\n {\n \"data\": {\n- \"image/png\": \"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\",\n+ \"image/png\": \"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\",\n \"text/plain\": [\n \"\"\n ]\n },\n \"metadata\": {},\n \"output_type\": \"display_data\"\n }\n"}]}]}, {"source1": "./usr/share/doc/python3-symfit/html/examples/ex_tikhonov.html", "source2": "./usr/share/doc/python3-symfit/html/examples/ex_tikhonov.html", "unified_diff": "@@ -53,15 +53,15 @@\n
\n
\n
\n
\n
\n
\n
\n-Matplotlib created a temporary cache directory at /tmp/matplotlib-tdvvp3jt because the default path (/nonexistent/first-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-bqkesga7 because the default path (/nonexistent/second-build/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n
\n
\n
Say \\(f(t) = t * exp(- t)\\), and \\(F(s)\\) is the Laplace transform of \\(f(t)\\). Let us first evaluate this transform using sympy.
\n
\n
[2]:\n
\n
\n@@ -105,15 +105,15 @@\n
\n
\n
[3]:\n
\n
\n
\n
\n-<matplotlib.legend.Legend at 0xe9e0df18>\n+<matplotlib.legend.Legend at 0xe9b10d38>\n
\n
\n
\n
\n
\n
\n \n@@ -206,16 +206,16 @@\n
\n
\n \n Parameter Value Standard Deviation\n a 5.449374e-02 None\n Status message Optimization terminated successfully.\n Number of iterations 14\n-Objective <symfit.core.objectives.LeastSquares object at 0xe9e7dea0>\n-Minimizer <symfit.core.minimizers.BFGS object at 0xf0c4c270>\n+Objective <symfit.core.objectives.LeastSquares object at 0xe97d1a50>\n+Minimizer <symfit.core.minimizers.BFGS object at 0xe97d1c90>\n \n Goodness of fit qualifiers:\n chi_squared 3.272835427084002e-19\n objective_value 1.636417713542001e-19\n r_squared -inf\n
\n
\n@@ -256,15 +256,15 @@\n
\n
\n
[8]:\n
\n
\n
\n
\n-<matplotlib.legend.Legend at 0xe99797c8>\n+<matplotlib.legend.Legend at 0xe9948ea0>\n
\n
\n
\n
\n
\n
\n \n@@ -292,15 +292,15 @@\n
\n
\n
[9]:\n
\n
\n
\n
\n-<matplotlib.legend.Legend at 0xe9a64480>\n+<matplotlib.legend.Legend at 0xe9782bd0>\n
![\"../_images/examples_ex_mexican_hat_5_1.png\"](\"../_images/examples_ex_mexican_hat_5_1.png\")
\n@@ -169,15 +169,15 @@\n ![\"../_images/examples_ex_tikhonov_5_1.png\"](\"../_images/examples_ex_tikhonov_5_1.png\")
\n@@ -206,16 +206,16 @@\n ![\"../_images/examples_ex_tikhonov_15_2.png\"](\"../_images/examples_ex_tikhonov_15_2.png\")
\n@@ -292,15 +292,15 @@\n ![\"../_images/examples_ex_tikhonov_17_1.png\"](\"../_images/examples_ex_tikhonov_17_1.png\")
\n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -7,17 +7,17 @@\n from symfit import (\n variables, parameters, Model, Fit, exp, laplace_transform, symbols,\n MatrixSymbol, sqrt, Inverse, CallableModel\n )\n import numpy as np\n import matplotlib.pyplot as plt\n \n-Matplotlib created a temporary cache directory at /tmp/matplotlib-tdvvp3jt\n-because the default path (/nonexistent/first-build/.config/matplotlib) is not a\n-writable directory; it is highly recommended to set the MPLCONFIGDIR\n+Matplotlib created a temporary cache directory at /tmp/matplotlib-bqkesga7\n+because the default path (/nonexistent/second-build/.config/matplotlib) is not\n+a writable directory; it is highly recommended to set the MPLCONFIGDIR\n environment variable to a writable directory, in particular to speed up the\n import of Matplotlib and to better support multiprocessing.\n Say \\(f(t) = t * exp(- t)\\), and \\(F(s)\\) is the Laplace transform of \\(f(t)\\).\n Let us first evaluate this transform using sympy.\n [2]:\n t, f, s, F = variables('t, f, s, F')\n model = Model({f: t * exp(- t)})\n@@ -45,15 +45,15 @@\n = F(s)$')\n plt.xlabel(r'$s_i$')\n plt.ylabel(r'$F(s_i)$')\n plt.xlim(0, None)\n plt.legend()\n \n [3]:\n-