Merge pull request #26 from CloudChaoszero/replace-pymc3-arviz-plots_part2

Replacing PyMC3 plots w/ Arviz plots & sigma Param change [Part 2]

Author: twiecki

examples/case_studies/BEST.ipynb

@@ -320,7 +320,7 @@
     "    trace = pm.sample(ndraws, tune=nburn, discard_tuned_samples=True)\n",
     "\n",
     "# plot the traces\n",
-    "_ = pm.traceplot(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
+    "_ = az.plot_trace(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
     "\n",
     "# put the chains in an array (for later!)\n",
     "samples_pymc3 = np.vstack((trace[\"m\"], trace[\"c\"])).T"
@@ -616,7 +616,7 @@
     "    trace = pm.sample(ndraws, tune=nburn, discard_tuned_samples=True)\n",
     "\n",
     "# plot the traces\n",
-    "_ = pm.traceplot(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
+    "_ = az.plot_trace(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
     "\n",
     "# put the chains in an array (for later!)\n",
     "samples_pymc3_2 = np.vstack((trace[\"m\"], trace[\"c\"])).T"
@@ -644,12 +644,12 @@
     "    theta = tt.as_tensor_variable([m, c])\n",
     "\n",
     "    # use a Normal distribution\n",
-    "    pm.Normal(\"likelihood\", mu=(m * x + c), sd=sigma, observed=data)\n",
+    "    pm.Normal(\"likelihood\", mu=(m * x + c), sigma=sigma, observed=data)\n",
     "\n",
     "    trace = pm.sample(ndraws, tune=nburn, discard_tuned_samples=True)\n",
     "\n",
     "# plot the traces\n",
-    "_ = pm.traceplot(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
+    "_ = az.plot_trace(trace, lines={\"m\": mtrue, \"c\": ctrue})\n",
     "\n",
     "# put the chains in an array (for later!)\n",
     "samples_pymc3_3 = np.vstack((trace[\"m\"], trace[\"c\"])).T"
@@ -832,7 +832,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.2"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,

examples/case_studies/LKJ.ipynb

@@ -122,7 +122,7 @@
     "\n",
     "The classical `WinBUGS` implementation (more information [here](http://glau.ca/?p=340)):\n",
     "\n",
-    "```\n",
+    "```python\n",
     "model\n",
     "{\n",
     "   for (i in 1 : regions) {\n",
@@ -2717,7 +2717,7 @@
     }
    ],
    "source": [
-    "summary2 = pm.summary(infdata2)\n",
+    "summary2 = az.summary(infdata2)\n",
     "summary2[summary2[\"r_hat\"] > 1.05]"
    ]
   },
@@ -3004,7 +3004,7 @@
     "Note that in the node $\\phi \\sim \\mathcal{N}(0, [D_\\tau (I - \\alpha B)]^{-1})$, we are computing the log-likelihood for a multivariate Gaussian distribution, which might not scale well in high-dimensions. We can take advantage of the fact that the covariance matrix here $[D_\\tau (I - \\alpha B)]^{-1}$ is **sparse**, and there are faster ways to compute its log-likelihood. \n",
     "\n",
     "For example, a more efficient sparse representation of the CAR in `Stan`:\n",
-    "```\n",
+    "```python\n",
     "functions {\n",
     "  /**\n",
     "  * Return the log probability of a proper conditional autoregressive (CAR) prior \n",
@@ -3040,7 +3040,7 @@
     "                    - tau * (phit_D * phi - alpha * (phit_W * phi)));\n",
     "  }\n",
     "}\n",
-    "```\n",
+    "```python\n",
     "with the data transformed in the model:\n",
     "```\n",
     "transformed data {\n",
@@ -3500,7 +3500,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.3"
+   "version": "3.8.6"
   }
  },
  "nbformat": 4,

examples/case_studies/blackbox_external_likelihood.ipynb

@@ -320,7 +320,7 @@
    "outputs": [],
    "source": [
     "with pm.Model() as lgcp_model:\n",
-    "    mu = pm.Normal(\"mu\", sd=3)\n",
+    "    mu = pm.Normal(\"mu\", sigma=3)\n",
     "    rho = pm.Uniform(\"rho\", lower=25, upper=200)\n",
     "    cov_scale = pm.Exponential(\"cov_scale\", lam=1)\n",
     "\n",
@@ -601,9 +601,7 @@
   {
    "cell_type": "code",
    "execution_count": 14,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -773,7 +771,7 @@
     "n_centroids = centroids.shape[0]\n",
     "\n",
     "with pm.Model() as mark_model:\n",
-    "    mu = pm.Normal(\"mu\", sd=3)\n",
+    "    mu = pm.Normal(\"mu\", sigma=3)\n",
     "    rho = pm.Uniform(\"rho\", lower=25, upper=200)\n",
     "\n",
     "    cov_scale = pm.Exponential(\"scale\", lam=1)\n",
@@ -809,14 +807,14 @@
    "outputs": [],
    "source": [
     "with mark_model:\n",
-    "    alpha = pm.Normal(\"alpha\", sd=10.0)\n",
-    "    beta = pm.Normal(\"beta\", sd=5)\n",
+    "    alpha = pm.Normal(\"alpha\", sigma=10.0)\n",
+    "    beta = pm.Normal(\"beta\", sigma=5)\n",
     "    eps_sd = pm.HalfCauchy(\"eps_sd\", beta=1.0)\n",
     "\n",
     "    marks = pm.Normal(\n",
     "        \"marks\",\n",
     "        mu=alpha + beta * intensity[n_centroids::],\n",
-    "        sd=eps_sd,\n",
+    "        sigma=eps_sd,\n",
     "        shape=n,\n",
     "        observed=data[\"marks\"].values,\n",
     "    )"
@@ -1023,9 +1021,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python PyMC3 (Dev)",
    "language": "python",
-   "name": "python3"
+   "name": "pymc3-dev-py38"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1037,7 +1035,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.6"
   },
   "toc": {
    "base_numbering": 1,
@@ -1054,5 +1052,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

examples/case_studies/conditional-autoregressive-model.ipynb

@@ -1499,9 +1499,9 @@
  "metadata": {
   "anaconda-cloud": {},
   "kernelspec": {
-   "display_name": "pymc-dev",
+   "display_name": "Python (PyMC3 Dev)",
    "language": "python",
-   "name": "pymc-dev"
+   "name": "pymc3-dev"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1513,7 +1513,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.2"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,