This is a PARSeq model uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends.
This model is related to a CausalLM task.
Model config:
- name: par_seq_backbone
- trainable: True
- dtype: {'module': 'keras', 'class_name': 'DTypePolicy', 'config': {'name': 'float32'}, 'registered_name': None}
- image_encoder: {'module': 'keras_hub.src.models.vit.vit_backbone', 'class_name': 'ViTBackbone', 'config': {'name': 'encoder', 'trainable': True, 'dtype': {'module': 'keras', 'class_name': 'DTypePolicy', 'config': {'name': 'float32'}, 'registered_name': None}, 'image_shape': [32, 128, 3], 'patch_size': [4, 8], 'num_layers': 12, 'num_heads': 6, 'hidden_dim': 384, 'mlp_dim': 1536, 'dropout_rate': 0.0, 'attention_dropout': 0.0, 'layer_norm_epsilon': 1e-06, 'use_mha_bias': True, 'use_mlp_bias': True, 'use_class_token': False, 'use_patch_bias': True}, 'registered_name': 'keras_hub>ViTBackbone'}
- vocabulary_size: 97
- max_label_length: 25
- decoder_hidden_dim: 384
- num_decoder_layers: 1
- num_decoder_heads: 12
- decoder_mlp_dim: 1536
- dropout_rate: 0.1
- attention_dropout: 0.1
This model card has been generated automatically and should be completed by the model author. See Model Cards documentation for more information.
- Downloads last month
- 7