[PT] fix: Alf reviews for PR #805

>
>
Co-authored-by: Felipe Schiavon <felipeschiavondeoliveira@gmail.com>
Co-authored-by: Bernardo Lago <bernardolago@gmail.com>"

docs/pt/week03/03-1.md

@@ -392,7 +392,7 @@ Então, por que o Deep Learning deveria estar enraizado na ideia de que nosso mu
 <!--<div align="center">Figure 11. Simon Thorpe's model of visual information flow in the brain <div>
 -->
 
-<div align="center">Figura 11. Modelo de Simon Thorpe de fluxo de informações visuais no cérebro<div>
+<div align="center">Figura 11. Modelo de Simon Thorpe de fluxo de informações visuais no cérebro</div>
 
 <!--Signals pass from the retina to the LGN (helps with contrast enhancement, gate control, etc.), then to the V1 primary visual cortex, V2, V4, then to the inferotemporal cortex (PIT), which is the part of the brain where categories are defined. Observations from open-brain surgery showed that if you show a human a film, neurons in the PIT will fire only when they detect certain images -- such as Jennifer Aniston or a person's grandmother -- and nothing else. The neural firings are invariant to things such as position, size, illumination, your grandmother's orientation, what she's wearing, etc.
 -->
@@ -422,7 +422,7 @@ Um outro insight do cérebro humano vem de Gallant & Van Essen, cujo modelo do c
 <!--<div align="center">Figure 12. Gallen & Van Essen's model of dorsal & ventral pathways in the brain <div>
 -->
 
-<div align="center">Figura 12. Modelo de Gallen e Van Essen das vias dorsais e ventrais no cérebro <div>
+<div align="center">Figura 12. Modelo de Gallen e Van Essen das vias dorsais e ventrais no cérebro </div>
 
 <!--The right side shows the ventral pathway, which tells you what you're looking at, while the left side shows the dorsal pathway, which identifies locations, geometry, and motion. They seem fairly separate in the human (and primate) visual cortex (with a few interactions between them of course).
 -->
@@ -443,7 +443,7 @@ O lado direito mostra a via ventral, que indica o que você está olhando, enqua
 <!--<div align="center">Figure 13. Hubel & Weisel's experiments with visual stimuli in cat brains <div>
 -->
 
-<div align="center"> Figura 13. Experimentos de Hubel e Wiesel com estímulos visuais em cérebros de gatos <div>
+<div align="center"> Figura 13. Experimentos de Hubel e Wiesel com estímulos visuais em cérebros de gatos </div>
 
 <!--Hubel and Weisel experiments used electrodes to measure neural firings in cat brains in response to visual stimuli. They discovered that neurons in the V1 region are only sensitive to certain areas of a visual field (called "receptive fields"), and detect oriented edges in that area. For example, they demonstrated that if you showed the cat a vertical bar and start rotating it, at a particular angle the neuron will fire. Similarly, as the bar moves away from that angle, the activation of the neuron diminishes. These activation-selective neurons Hubel & Weisel named "simple cells", for their ability to detect local features.
 -->
@@ -474,7 +474,7 @@ Outro tipo de neurônio, que eles chamaram de "células complexas", agregam a sa
 <!--<div align="center">Figure 14. Fukushima's CNN model <div>
 -->
 
-<div align="center"> Figura 14. Modelo CNN de Fukushima <div>
+<div align="center"> Figura 14. Modelo CNN de Fukushima </div>
 
 <!--Fukushima was the first to implement the idea of multiple layers of simple cells and complex cells with computer models, using a dataset of handwritten digits. Some of these feature detectors were hand-crafted or learned, though the learning used unsupervised clustering algorithms, trained separately for each layer, as backpropagation was not yet in use.
 -->