fix notebooks' comments following @taku-y recommendations

Author: ferrine

docs/source/notebooks/convolutional_vae_keras_advi.ipynb

@@ -33,7 +33,9 @@
   {
    "cell_type": "code",
    "execution_count": 1,
-   "metadata": {},
+   "metadata": {
+    "collapsed": true
+   },
    "outputs": [],
    "source": [
     "#!pip install --upgrade keras\n",
@@ -107,7 +109,6 @@
     "from keras.layers import InputLayer, BatchNormalization, Dense, Conv2D, Deconv2D, Activation, Flatten, Reshape\n",
     "import numpy as np\n",
     "import pymc3 as pm\n",
-    "from pymc3.variational import advi_minibatch\n",
     "from theano import shared, config, function, clone, pp\n",
     "import theano.tensor as tt\n",
     "import keras\n",
@@ -189,7 +190,9 @@
   {
    "cell_type": "code",
    "execution_count": 5,
-   "metadata": {},
+   "metadata": {
+    "collapsed": true
+   },
    "outputs": [],
    "source": [
     "from keras.models import Sequential\n",
@@ -532,7 +535,9 @@
   {
    "cell_type": "code",
    "execution_count": 44,
-   "metadata": {},
+   "metadata": {
+    "collapsed": true
+   },
    "outputs": [],
    "source": [
     "with pm.Model() as model:\n",
@@ -758,7 +763,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.1"
+   "version": "3.6.0"
   },
   "latex_envs": {
    "bibliofile": "biblio.bib",

docs/source/notebooks/lda-advi-aevb.ipynb

@@ -231,7 +231,9 @@
   {
    "cell_type": "code",
    "execution_count": 6,
-   "metadata": {},
+   "metadata": {
+    "collapsed": true
+   },
    "outputs": [],
    "source": [
     "n_topics = 10\n",
@@ -256,7 +258,7 @@
    "metadata": {},
    "source": [
     "## Encoder\n",
-    "Given a document, the encoder calculates variational parameters of the (transformed) latent variables, more specifically, parameters of Gaussian distributions in the unconstrained real coordinate space. The `encode()` method is required to output variational means and stds as a tuple, as shown in the following code. As explained above, the number of variational parameters is `2 * (n_topics) - 1`. Specifically, the shape of `zs_mean` (or `zs_std`) in the method is `(minibatch_size, n_topics - 1)`. It should be noted that `zs_std` is defined as log-transformed standard deviation and this is automativally exponentiated (thus bounded to be positive) in `advi_minibatch()`, the estimation function. \n",
+    "Given a document, the encoder calculates variational parameters of the (transformed) latent variables, more specifically, parameters of Gaussian distributions in the unconstrained real coordinate space. The `encode()` method is required to output variational means and stds as a tuple, as shown in the following code. As explained above, the number of variational parameters is `2 * (n_topics) - 1`. Specifically, the shape of `zs_mean` (or `zs_std`) in the method is `(minibatch_size, n_topics - 1)`. It should be noted that `zs_std` is defined as $\\rho = log(exp(std) - 1)$ in `ADVI` and bounded to be positive. The inverse parametrization is $std = log(1+exp(\\rho))$ and considered to be numericaly stable.\n",
     "\n",
     "To enhance generalization ability to unseen words, a bernoulli corruption process is applied to the inputted documents. Unfortunately, I have never see any significant improvement with this. "
    ]
@@ -724,7 +726,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.1"
+   "version": "3.6.0"
   },
   "latex_envs": {
    "bibliofile": "biblio.bib",