Loading...
Loading
ConnitoAI
ვ (Vin)SN102Contributors train specialized expert modules that are aggregated into powerful AI systems, without massive centralized compute.
Loading
Contributors train specialized expert modules that are aggregated into powerful AI systems, without massive centralized compute.
This information was auto-extracted from the GitHub repository and may be incomplete or out of date. Always refer to the subnet's GitHub repository for the latest instructions.
8-bit AdamW optimizer (bitsandbytes) used to reduce memory footprint for fitting training steps on smaller GPUs
Install all dependencies including PyTorch, transformers, and distributed training libraries
pip install -r requirements.txt
Install CUDA-optimized PyTorch wheel (install from pytorch.org first as noted in requirements)
pip install torch==2.10.0 --index-url https://download.pytorch.org/whl/cu118
Run miner using the console entry point
weightnet-miner
Install all dependencies including FastAPI, bittensor, and observability tools
pip install -r requirements.txt
Install CUDA-optimized PyTorch wheel
pip install torch==2.10.0 --index-url https://download.pytorch.org/whl/cu118
Run validator using the console entry point
weightnet-validator