Upload Nicheformer model

README.md

@@ -29,33 +29,41 @@ Nicheformer is built on a transformer architecture with the following key featur
 
 ```python
 from transformers import AutoModelForMaskedLM, AutoTokenizer
+import anndata as ad
 
 # Load model and tokenizer
-model = AutoModelForMaskedLM.from_pretrained("your-username/nicheformer")
-tokenizer = AutoTokenizer.from_pretrained("your-username/nicheformer")
+model = AutoModelForMaskedLM.from_pretrained("aletlvl/Nicheformer")
+tokenizer = AutoTokenizer.from_pretrained("aletlvl/Nicheformer")
 
-# Example 1: Manual masking
-masked_text = "The [MASK] cell is an important immune cell type."
-inputs = tokenizer(masked_text, return_tensors="pt")
-outputs = model(**inputs)
+# Load your single-cell data
+adata = ad.read_h5ad("your_data.h5ad")
+
+# Tokenize the data
+inputs = tokenizer(adata)
 
-# Example 2: Automatic masking (typically used during training)
-text = "The T cell is an important immune cell type."
-inputs = tokenizer(text, return_tensors="pt")
-outputs = model(**inputs, apply_masking=True)  # This will automatically mask tokens
+# Get predictions
+outputs = model(**inputs)
 ```
 
 ## Training Data
 
-[Describe the training data used for the model]
-
-## Evaluation Results
+The model was trained on single-cell gene expression data from various tissues and organisms. It supports:
 
-[Include evaluation metrics and results]
+- **Modalities**: spatial and dissociated
+- **Species**: human and mouse
+- **Technologies**: 
+  - MERFISH
+  - CosMx
+  - Xenium
+  - 10x Genomics (various versions)
+  - CITE-seq
+  - Smart-seq v4
 
 ## Limitations
 
-[Describe any known limitations or biases of the model]
+- The model is specifically designed for gene expression data and may not generalize to other types of biological data
+- Performance may vary depending on the quality and type of input data
+- The model works best with data from supported species and technologies
 
 ## Citation
 
@@ -67,8 +75,58 @@ If you use this model in your research, please cite:
 
 ## License
 
-This model is released under [specify license]
+This model is released under the MIT License. See the LICENSE file for more details.
 
 ## Contact
 
-[Add contact information for questions and issues] 
+For questions and issues, please open an issue on the GitHub repository or contact the maintainers.
+
+# nicheformer
+
+This is the official repository for **Nicheformer: a foundation model for single-cell and spatial omics**
+
+[![Preprint](https://img.shields.io/badge/preprint-available-brightgreen)](https://www.biorxiv.org/content/10.1101/2024.04.15.589472v1)  
+
+A rendered Jupyter book version of this repository will be available soon.
+
+## Citation
+
+If you use our tool or build upon our concepts in your own work, please cite it as
+
+```
+Schaar, A.C., Tejada-Lapuerta, A., et al. Nicheformer: a foundation model for single-cell and spatial omics. bioRxiv (2024). doi: https://doi.org/10.1101/2024.04.15.589472
+```
+
+## Installation
+
+You need to have Python 3.9 or newer installed on your system. If you don't have
+Python installed, we recommend installing [Mambaforge](https://github.com/conda-forge/miniforge#mambaforge).
+
+
+<!--
+1) Install the latest release of `nicheformer` from `PyPI <https://pypi.org/project/nicheformer/>`_:
+
+```bash
+pip install nicheformer
+```
+-->
+
+Install the latest development version:
+
+```bash
+git clone https://github.com/theislab/nicheformer.git
+cd nicheformer
+pip install -e .
+```
+## Nicheformer data
+We provide examplary data loading scripts in the data subdirectory that can be used as templates for loading the spatial omics datasets and datasets retreived from GEO. 
+
+## Pretraining weights
+We provide the Nicheformer pretraining weights on Mendeley data, they can be downloaded from [here](https://data.mendeley.com/preview/87gm9hrgm8?a=d95a6dde-e054-4245-a7eb-0522d6ea7dff). 
+
+## Contact
+
+For questions and help requests, you can reach out (preferably) on GitHub or email to the corresponding author. 
+
+
+[issue-tracker]: https://github.com/theislab/nicheformer/issues