You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+1-29Lines changed: 1 addition & 29 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -65,37 +65,9 @@ We get the shortest paths from the source node 0 to all other nodes in the graph
65
65
pip install edsger
66
66
```
67
67
68
-
### Windows Performance Optimization
69
-
70
-
For optimal performance on Windows, we recommend using the same GCC toolchain as SciPy:
71
-
72
-
#### Option 1: GCC Toolchain (Recommended)
73
-
```bash
74
-
# Install MinGW-w64 (see INSTALL_MINGW_WINDOWS.md for details)
75
-
conda install -c conda-forge m2w64-toolchain
76
-
77
-
# Build with GCC (same as SciPy)
78
-
git clone https://github.com/aetperf/edsger.git
79
-
cd edsger
80
-
pip install -e .
81
-
```
82
-
83
-
#### Option 2: MSVC Optimizations (Fallback)
84
-
```bash
85
-
git clone https://github.com/aetperf/edsger.git
86
-
cd edsger
87
-
pip install -e .
88
-
```
89
-
90
-
The build system automatically detects the best available compiler and applies appropriate optimizations. For more details, see [WINDOWS_OPTIMIZATION.md](WINDOWS_OPTIMIZATION.md).
91
-
92
-
**Performance improvement**:
93
-
-**With GCC**: Target 1.4-1.6x slower than Linux (vs current 2.0x)
94
-
-**With MSVC**: Up to 50% faster than unoptimized builds
95
-
96
68
## Why Use Edsger?
97
69
98
-
Edsger is designed to be **dataframe-friendly**, providing seamless integration with pandas workflows for graph algorithms. Also it is rather efficient. Our benchmarks on the USA road network (23.9M vertices, 57.7M edges) demonstrate nice performance:
70
+
Edsger is designed to be **dataframe-friendly**, providing seamless integration with pandas workflows for graph algorithms. Also it is rather efficient on Linux. Our benchmarks on the USA road network (23.9M vertices, 57.7M edges) demonstrate nice performance:
0 commit comments