Memoraith is a cutting-edge, lightweight model profiler for deep learning frameworks, providing unparalleled insights into neural network performance. Developed with precision and efficiency in mind, it helps developers and researchers optimize their models through detailed performance analysis.
-
🔍 Advanced Profiling
- High-precision memory tracking (CPU & GPU)
- Microsecond-accurate computation timing
- Layer-by-layer performance analysis
-
🎯 Intelligent Analysis
- Sophisticated bottleneck detection
- Anomaly identification
- Optimization recommendations
-
📊 Rich Visualization
- Interactive dashboards
- Real-time monitoring
- Comprehensive reports
-
🛠 Framework Support
- PyTorch integration
- TensorFlow support
- Extensible architecture
Basic installation:
pip install memoraithFull installation with GPU support and extra features:
pip install memoraith[full]Here's a simple example using PyTorch:
from memoraith import profile_model, set_output_path
import torch
import torch.nn as nn
# Set output directory for profiling results
set_output_path('profiling_results/')
# Define your model
class AdvancedNet(nn.Module):
def __init__(self):
super(AdvancedNet, self).__init__()
self.conv1 = nn.Conv2d(3, 64, kernel_size=3, padding=1)
self.conv2 = nn.Conv2d(64, 128, kernel_size=3, padding=1)
self.fc = nn.Linear(128 * 8 * 8, 10)
def forward(self, x):
x = torch.relu(self.conv1(x))
x = torch.relu(self.conv2(x))
x = x.view(x.size(0), -1)
return self.fc(x)
# Add profiling decorator
@profile_model(memory=True, computation=True, gpu=True)
def train_model(model):
optimizer = torch.optim.Adam(model.parameters())
for _ in range(100):
input_data = torch.randn(32, 3, 32, 32)
output = model(input_data)
loss = output.sum()
loss.backward()
optimizer.step()
if __name__ == "__main__":
model = AdvancedNet()
train_model(model)Visit our comprehensive documentation for:
- Detailed API reference
- Advanced usage examples
- Best practices
- Troubleshooting guides
We welcome contributions! See our Contributing Guide for:
- Code of conduct
- Development setup
- Submission guidelines
- Testing procedures
Memoraith is released under the MIT License. See LICENSE file for details.
Need help?
If you use Memoraith in your research, please cite:
@software{memoraith,
author = {El Jouhfi, Mehdi},
title = {Memoraith: Advanced Lightweight Model Profiler for Deep Learning},
year = {2024},
url = {https://github.com/mehdi342/Memoraith},
version = {0.5.0}
}For inquiries, reach out to Mehdi El Jouhfi