Instructions to use vikenkd/vqa-llm1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vikenkd/vqa-llm1 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("vikenkd/vqa-llm1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "batchnorm_layer": 256, | |
| "dropout": 0.3, | |
| "embed_dim": 512, | |
| "expansion_factor": 4, | |
| "hidden_dim": 128, | |
| "input_layer": 128, | |
| "model_type": "visualqa", | |
| "n_dense": 6, | |
| "n_heads": 8, | |
| "num_layers": 2, | |
| "output_classification": 3454, | |
| "output_layer": 256, | |
| "transformers_version": "4.41.2", | |
| "w_s": 0.5 | |
| } | |