Merge pull request #148 from linok-bc/master

add snemi benchmark

Author: linok-bc

docs/source/notes/installation.rst

@@ -17,9 +17,9 @@ Please follow the steps below for a successful installation:
 
 .. code-block:: bash
 
-    conda create -n pytc python=3.9
+    conda create -y -n pytc python=3.9
     conda activate pytc
-    conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.3 mkl=2024.0.0 -c pytorch
+    conda install -y pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit==11.3.1 -c pytorch
 
 More options to install PyTorch can be found `here <https://pytorch.org/get-started/locally/>`_. Our package has been tested with CUDA 10.2 and 11.4. Then please ensure that at least PyTorch **1.10.0** is installed:
 
@@ -31,12 +31,15 @@ More options to install PyTorch can be found `here <https://pytorch.org/get-star
 2 - Install PyTorch Connectomics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-.. code-block:: none
+.. code-block:: bash
 
     git clone https://github.com/zudi-lin/pytorch_connectomics.git
     cd pytorch_connectomics
     pip install --editable .
 
+    # fix some dependency issues
+    pip install pillow==9.4.0
+
 We install the package in editable mode by default so that there is no need to
 re-install it when making changes to the code. 
 

docs/source/tutorials/neuron.rst

@@ -30,6 +30,9 @@ be regarded as a more informative version of boundary map as it contains the aff
 The figure above shows examples of EM images, segmentation and affinity map from the SNEMI3D dataset. Since the
 3D affinity map has 3 channels, we can visualize them as RGB images.
 
+
+    .. tip:: A notebook for visualizing results is provided for users in the `Github repository <https://github.com/zudi-lin/pytorch_connectomics/tree/master/notebooks/tutorial_benchmarks/snemi_benchmark.ipynb>`_. Users are able to download this notebook and produce evaluation results using a pretrained benchmark. Due to incompatiability with Colab and neuroglancer, it is recommended that users utilize a personal computer/HPC with at least 12GB of video RAM.
+
 1 - Get the data
 ^^^^^^^^^^^^^^^^^^
 
@@ -54,6 +57,7 @@ For description of the SNEMI dataset please check `this page <https://vcg.github
 Provide the **YAML** configuration files to run training:
 
 .. code-block:: none
+   
    source activate py3_torch
    python -u scripts/main.py \
    --config-base configs/SNEMI/SNEMI-Base.yaml \