How uliontse/mlgb works
MLGB is a TensorFlow/PyTorch library for CTR prediction and recommendation. The main workflow is selecting a model through the public API, preparing feature inputs, and running ranking, matching, or multitask model implementations. Both framework stacks provide parallel input, component, and model layers; matching also has retrieval-related components. The README excerpt is truncated, so detailed capabilities are represented at the documented model-family level. Edges rely on the README or SOURCE INDEX references; unsampled implementation details are not inferred.
Main entry points: tf/agg.py TF model registry · ranking.py Ranking models · mtl.py Multitask models · torch/agg.py Torch model registry
Components 21
API and data
- Model API
mlgb/main.py - Data utilities
mlgb/data.py - Feature engineering
mlgb/fe.py
TensorFlow
- TF model registry
mlgb/tf/agg.py - Feature inputs
mlgb/tf/inputs.py - Model components
mlgb/tf/components - Ranking models
mlgb/tf/models/ranking.py - Matching models
mlgb/tf/models/matching.py - Multitask models
mlgb/tf/models/mtl.py - Ranking modules
mlgb/tf/modules/ranking - Matching modules
mlgb/tf/modules/matching.py
PyTorch
- Torch model registry
mlgb/torch/agg.py - Feature inputs
mlgb/torch/inputs.py - Model components
mlgb/torch/components - Ranking models
mlgb/torch/models/ranking.py - Matching models
mlgb/torch/models/matching.py - Multitask models
mlgb/torch/models/mtl.py - Ranking modules
mlgb/torch/modules/ranking - Matching modules
mlgb/torch/modules/matching.py
Shared support
- Utility helpers
mlgb/utils.py
Other components
- Library user
Connections 27
- Library user → Model API: requests model
- Model API → TF model registry: dispatches TF
mlgb/main.py - Model API → Torch model registry: dispatches Torch
mlgb/main.py - Data utilities → Feature engineering: uses features
mlgb/data.py - Data utilities → Utility helpers: uses helpers
mlgb/data.py - TF model registry → Ranking models: exposes ranking
mlgb/tf/agg.py - TF model registry → Matching models: exposes matching
mlgb/tf/agg.py - TF model registry → Multitask models: exposes multitask
mlgb/tf/agg.py - Ranking models → Feature inputs: uses inputs
mlgb/tf/models/ranking.py - Ranking models → Model components: uses components
mlgb/tf/models/ranking.py - Ranking models → Ranking modules: uses modules
mlgb/tf/models/ranking.py - Matching models → Model components: uses components
mlgb/tf/models/matching.py - Matching models → Matching modules: uses modules
mlgb/tf/models/matching.py - Multitask models → Feature inputs: uses inputs
mlgb/tf/models/mtl.py - Multitask models → Model components: uses components
mlgb/tf/models/mtl.py - Multitask models → Ranking modules: uses multitask module
mlgb/tf/models/mtl.py - Torch model registry → Ranking models: exposes ranking
mlgb/torch/agg.py - Torch model registry → Matching models: exposes matching
mlgb/torch/agg.py - Torch model registry → Multitask models: exposes multitask
mlgb/torch/agg.py - Ranking models → Feature inputs: uses inputs
mlgb/torch/models/ranking.py - Ranking models → Model components: uses components
mlgb/torch/models/ranking.py - Ranking models → Ranking modules: uses modules
mlgb/torch/models/ranking.py - Matching models → Model components: uses components
mlgb/torch/models/matching.py - Matching models → Matching modules: uses modules
mlgb/torch/models/matching.py - Multitask models → Feature inputs: uses inputs
mlgb/torch/models/mtl.py - Multitask models → Model components: uses components
mlgb/torch/models/mtl.py - Multitask models → Ranking modules: uses multitask module
mlgb/torch/models/mtl.py