change batch size and README file

README.md

@@ -8,7 +8,7 @@ The analysis can be performed by HEIG (will have more in the near future):
 - Cross-trait genetic correlation between voxels and non-imaging phenotypes
 
 ## Version
-- [v1.0.0](https://github.com/Zhiwen-Owen-Jiang/heig/releases/tag/v1.0.0): The initial version of HEIG.
+- [v1.1.0](https://github.com/Zhiwen-Owen-Jiang/heig/releases/tag/v1.1.0): Support multi-threading computation; many changes in data format; not compatible with v1.0.0.
 
 ## System Requirements
 ### OS Requirements
@@ -17,14 +17,14 @@ HEIG is supported for macOS and Linux. It has been tested on the following syste
 - MacOS Sonoma 14.5
 
 ### Python Dependencies
-We implement HEIG on Python 3.11. Specific package dependencies are provided in [requirements](https://github.com/Zhiwen-Owen-Jiang/heig/blob/pub/requirements.txt).
+HEIG is implemented in Python 3.11. Specific package dependencies are provided in [requirements](https://github.com/Zhiwen-Owen-Jiang/heig/blob/pub/requirements.txt).
 
 ## Getting Started
-First download the [released version](https://github.com/Zhiwen-Owen-Jiang/heig/releases/tag/v1.0.0), unzip it, and navigate to the extracted folder:
+First download the [released version](https://github.com/Zhiwen-Owen-Jiang/heig/releases/tag/v1.1.0), unzip it, and navigate to the extracted folder:
 ```
-wget -O heig-1.0.0.zip  https://github.com/Zhiwen-Owen-Jiang/heig/archive/refs/tags/v1.0.0.zip
-unzip heig-1.0.0.zip
-cd heig-1.0.0
+wget -O heig-1.1.0.zip  https://github.com/Zhiwen-Owen-Jiang/heig/archive/refs/tags/v1.1.0.zip
+unzip heig-1.1.0.zip
+cd heig-1.1.0
 ```
 Install [Miniconda](https://docs.anaconda.com/free/miniconda/miniconda-install/) based on your OS, and build the environment and install all dependencies for HEIG. This step may take ~5 minutes.
 ```
@@ -34,7 +34,7 @@ conda activate heig
 Or you can do it manually. 
 
 ## How to use HEIG
-We provided detailed [tutorial](https://github.com/Zhiwen-Owen-Jiang/heig/wiki) for using HEIG. The example data used in the tutorial can be downloaded [here](https://zenodo.org/records/11075259). Common issues are described in the [FAQ](https://github.com/Zhiwen-Owen-Jiang/heig/wiki/FAQ).
+We provided detailed [tutorial](https://github.com/Zhiwen-Owen-Jiang/heig/wiki) for using HEIG. The example data used in the tutorial for v1.1.0 can be downloaded [here](https://zenodo.org/records/13770930). Common issues are described in the [FAQ](https://github.com/Zhiwen-Owen-Jiang/heig/wiki/FAQ).
 
 If that does not work, email Owen Jiang <[email protected]> or <[email protected]>.
 

heig/fpca.py

@@ -410,9 +410,9 @@ def _get_batch_size(self, max_n_pc, n_sub):
             else:
                 batch_size = n_sub // (n_sub // 50000 + 1)
         else:
-            if self.n_top > 15000 or n_sub > 50000:
+            if self.n_top > 2000 or n_sub > 20000:
                 i = 2
-                while n_sub // i > 50000:
+                while n_sub // i > 20000:
                     i += 1
                 batch_size = n_sub // i
             else: