docs: add logo, fix typo

Author: okunator

README.md

@@ -1,7 +1,9 @@
-# cellseg_models.pytorch
-
 <div align="center">
 
+![Logo](./images/logo.png)
+
+**Python library for 2D cell/nuclei instance segmentation models written with [PyTorch](https://pytorch.org/).**
+
 [![Generic badge](https://img.shields.io/badge/License-MIT-<COLOR>.svg?style=for-the-badge)](https://github.com/okunator/cellseg_models.pytorch/blob/master/LICENSE)
 [![PyTorch - Version](https://img.shields.io/badge/PYTORCH-1.8+-red?style=for-the-badge&logo=pytorch)](https://pytorch.org/)
 [![Python - Version](https://img.shields.io/badge/PYTHON-3.7+-red?style=for-the-badge&logo=python&logoColor=white)](https://www.python.org/)
@@ -16,8 +18,6 @@
 
 <div align="center">
 
-**Python library for 2D cell/nuclei instance segmentation models written with [PyTorch](https://pytorch.org/).**
-
 </div>
 
 ## Introduction

examples/pannuke_nuclei_segmentation_stardist.ipynb

@@ -275,7 +275,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Typically, encoder-decoder based nuclei segmentation model outputs require post-processing. The main-task of the post-processing is to separate clumped nuclear-objects which is a renowned problem in nuclei segmentation. With `cellseg_models.pytorch`, inference and post-processing can be executed with specific `Inferer` classes that can be found in the `csmp.inference` module. Since the Pannuke-dataset has only 256x256px images, we can use the `ResizeInferer` to run the inference and post-processing (without actually resizing the images). The `Inferers` take in an input directory and a set of arguments, from which, the `instance_postproc` is the most important since it sets the post-procesing method to be used. Here, naturally, we will use `stardist` post-processing since we're running inference for a Stardist model.\n",
+    "Typically, encoder-decoder based nuclei segmentation model outputs require post-processing. The main-task of the post-processing is to separate clumped nuclear-objects which is a renowned problem in nuclei segmentation. With `cellseg_models.pytorch`, inference and post-processing can be executed with specific `Inferer` classes that can be found in the `csmp.inference` module. Since the Pannuke-dataset has only 256x256px images, we can use the `ResizeInferer` to run the inference and post-processing (without actually resizing the images). The `Inferers` take in an input directory and a set of arguments, from which, the `instance_postproc` is the most important since it sets the post-processing method to be used. Here, naturally, we will use `stardist` post-processing since we're running inference for a Stardist model.\n",
     "\n",
     "Other important params include: \n",
     "- `out_activations` - Sets the output activation functions for each of the model outputs\n",