pvt_v2_b2

Converted TIMM image classification model for LiteRT.

• Source architecture: pvt_v2_b2
• Source checkpoint: timm/pvt_v2_b2.in1k
• File: model.tflite
• Input: float32 tensor in NCHW layout, shape [1, 3, 224, 224]
• Output: ImageNet-1K logits, shape [1, 1000]

Runtime Status

• CPU smoke test: passed with LiteRT CompiledModel.
• GPU delegation: currently blocked for this model by rank-5 tensor patterns in the GPU backend, mostly RESHAPE, TRANSPOSE, and related window/attention operations. The model is published as CPU-ready while GPU support is being improved.

Model Details

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 25.4
  • GMACs: 4.0
  • Activations (M): 27.5
  • Image size: 224 x 224
• Papers:
  • PVT v2: Improved Baselines with Pyramid Vision Transformer: https://arxiv.org/abs/2106.13797
• Dataset: ImageNet-1k
• Original: https://github.com/whai362/PVT

Citation

@article{wang2021pvtv2,
  title={Pvtv2: Improved baselines with pyramid vision transformer},
  author={Wang, Wenhai and Xie, Enze and Li, Xiang and Fan, Deng-Ping and Song, Kaitao and Liang, Ding and Lu, Tong and Luo, Ping and Shao, Ling},
  journal={Computational Visual Media},
  volume={8},
  number={3},
  pages={1--10},
  year={2022},
  publisher={Springer}
}