How hzlbbfrog/crackmamba works
The checked-in runtime is the vendored Mamba 1.2.0 language-model library; the README describes CrackMamba crack segmentation, but this tree contains no segmentation model or image-to-mask workflow. The represented runtime configures a Mamba language model, runs its selective state-space operations, and supports token generation and Hugging Face checkpoint loading. The README’s installation and Jetson tutorials are not runtime components. Connections are limited to relationships visible in excerpts or the source index.
Main entry points: mixer_seq_simple.py LM model · mamba_simple.py Mamba block · selective_scan_interface.py Scan interface · generation.py Token generation
Components 13
Model assembly
- Model config
mamba-1.2.0/mamba_ssm/models/config_mamba.py - LM model
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - Mamba block
mamba-1.2.0/mamba_ssm/modules/mamba_simple.py
State-space operations
- Triton norms
mamba-1.2.0/mamba_ssm/ops/triton/layernorm.py - Scan interface
mamba-1.2.0/mamba_ssm/ops/selective_scan_interface.py - State update
mamba-1.2.0/mamba_ssm/ops/triton/selective_state_update.py
Inference support
- Token generation
mamba-1.2.0/mamba_ssm/utils/generation.py - HF checkpoint loading
mamba-1.2.0/mamba_ssm/utils/hf.py - LM evaluation adapter
mamba-1.2.0/evals/lm_harness_eval.py
CUDA extension
- Scan bindings
mamba-1.2.0/csrc/selective_scan/selective_scan.cpp - Scan parameters
mamba-1.2.0/csrc/selective_scan/selective_scan.h
Other components
- Model user
- Hugging Face Hub
Connections 14
- Model user → LM model: uses model
- LM model → Model config: configures from
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - LM model → Mamba block: builds blocks
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - LM model → Triton norms: uses norms
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - Mamba block → Scan interface: runs scan
mamba-1.2.0/mamba_ssm/modules/mamba_simple.py - Mamba block → State update: updates state
mamba-1.2.0/mamba_ssm/modules/mamba_simple.py - Mamba block → Triton norms: uses norms
mamba-1.2.0/mamba_ssm/modules/mamba_simple.py - Scan interface → Scan bindings: calls extension
mamba-1.2.0/mamba_ssm/ops/selective_scan_interface.py - Scan bindings → Scan parameters: uses parameter types
mamba-1.2.0/csrc/selective_scan/selective_scan.cpp - LM model → Token generation: inherits generation
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - Token generation → LM model: runs model
mamba-1.2.0/mamba_ssm/utils/generation.py - LM model → HF checkpoint loading: loads checkpoints
mamba-1.2.0/mamba_ssm/models/mixer_seq_simple.py - HF checkpoint loading → Hugging Face Hub: fetches config and weights
mamba-1.2.0/mamba_ssm/utils/hf.py - LM evaluation adapter → LM model: loads pretrained model
mamba-1.2.0/evals/lm_harness_eval.py