From 6981cfc2409a8d7881f71a04524f8a22e88e83f8 Mon Sep 17 00:00:00 2001
From: Minxing Pang <32996899+maxpmx@users.noreply.github.com>
Date: Sun, 27 Oct 2024 14:15:00 -0400
Subject: [PATCH] update

---
 README.md                  |  2 +-
 docs/source/index.rst      |  1 +
 docs/source/quickstart.rst | 21 +++++++++++++++------
 3 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/README.md b/README.md
index 569f6ce..5f5cc5c 100644
--- a/README.md
+++ b/README.md
@@ -56,7 +56,7 @@ cd CelloType
 Then Download the model weights:
 
 ```bash
-cd data
+cd models
 sh download.sh
 cd ..
 ```
diff --git a/docs/source/index.rst b/docs/source/index.rst
index 7713b73..f328d1f 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -15,6 +15,7 @@ CelloType is an end-to-end Transformer-based method for automated cell/nucleus s
    :caption: Examples:
 
    notebooks/cell_segmentation
+   inference_large
    notebooks/tissue_annotation
    
 .. toctree::
diff --git a/docs/source/quickstart.rst b/docs/source/quickstart.rst
index 4bc8d6a..d0ac278 100644
--- a/docs/source/quickstart.rst
+++ b/docs/source/quickstart.rst
@@ -1,22 +1,31 @@
 Quickstart
 ------------------------------
 
-Clone the repository:
+Clone the repository and download the pre-trained models:
+~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: bash
 
     git clone https://github.com/maxpmx/CelloType.git
     cd CelloType
 
-Then Download the model weights:
-
 .. code-block:: bash
 
-    cd data
+    cd models
     sh download.sh
     cd ..
 
-Predict on an example image:
+Prepare the input images
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Convert the input images into an RGB format where the blue channel represents the nuclear channel, the green channel corresponds to the membrane channel.
+
+.. image:: ../../data/example/example_tissuenet.png
+    :width: 250px
+    :alt: drawing
+
+Inference the cell segmentation
+~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: python
 
@@ -31,4 +40,4 @@ Predict on an example image:
       device='cuda', 
       config_path='./configs/maskdino_R50_bs16_50ep_4s_dowsample1_2048.yaml')
 
-    mask = model.predict(img) # [H, W]
\ No newline at end of file
+    mask = model.predict(img) # [H, W]