Update license information in model card
Browse files
README.md
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
---
|
| 2 |
-
license:
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
base_model: google/gemma-2-2b
|
|
@@ -8,8 +8,16 @@ pipeline_tag: text-generation
|
|
| 8 |
tags:
|
| 9 |
- biology
|
| 10 |
- scRNAseq
|
| 11 |
-
-
|
| 12 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
---
|
| 14 |
|
| 15 |
# C2S-Scale-Gemma-2B model card
|
|
@@ -51,7 +59,7 @@ classification, and generating biologically meaningful cell representations.
|
|
| 51 |
|
| 52 |
* Versatility: Demonstrates strong performance across a diverse set of single-cell and multi-cell tasks.
|
| 53 |
* Scalability: Trained on a massive dataset of over 57 million cells, showcasing the power of scaling LLMs for biological data.
|
| 54 |
-
* Generative Power: Capable of generating realistic single-cell gene expression profiles
|
| 55 |
* Foundation for Fine-tuning: Can serve as a powerful pretrained foundation for specialized, domain-specific single-cell analysis tasks.
|
| 56 |
|
| 57 |
**Potential Applications**
|
|
@@ -113,7 +121,7 @@ model = AutoModelForCausalLM.from_pretrained(
|
|
| 113 |
).to(device)
|
| 114 |
|
| 115 |
# Format prompt (see previous section)
|
| 116 |
-
cell_sentence = "MALAT1 TMSB4X B2M EEF1A1 H3F3B ACTB FTL RPL13 ..." # Truncated for example
|
| 117 |
num_genes = 1000
|
| 118 |
organism = "Homo sapiens"
|
| 119 |
|
|
@@ -183,8 +191,7 @@ established best practices for splitting data to ensure robust and unbiased asse
|
|
| 183 |
|
| 184 |
## License
|
| 185 |
|
| 186 |
-
The model weights
|
| 187 |
-
The underlying codebase for the Cell2Sentence project is licensed under CC BY-NC-ND 4.0.
|
| 188 |
|
| 189 |
## Implementation information
|
| 190 |
|
|
@@ -241,4 +248,4 @@ C2S-Scale provides a powerful, versatile, and scalable tool for single-cell anal
|
|
| 241 |
# Gemma-2 Links
|
| 242 |
- HuggingFace: https://huggingface.co/google/gemma-2-2b
|
| 243 |
- Gemma-2 Blog Post: [Gemma explained: What's new in Gemma 2](https://developers.googleblog.com/en/gemma-explained-new-in-gemma-2/)
|
| 244 |
-
- Technical report: https://storage.googleapis.com/deepmind-media/gemma/gemma-2-report.pdf
|
|
|
|
| 1 |
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
base_model: google/gemma-2-2b
|
|
|
|
| 8 |
tags:
|
| 9 |
- biology
|
| 10 |
- scRNAseq
|
| 11 |
+
- Gemma-2
|
| 12 |
+
- genomics
|
| 13 |
+
- computational-biology
|
| 14 |
+
- bioinformatics
|
| 15 |
+
- gene-expression
|
| 16 |
+
- cell-biology
|
| 17 |
+
- transformers
|
| 18 |
+
- pytorch
|
| 19 |
+
- cell-type-annotation
|
| 20 |
+
- Question Answering
|
| 21 |
---
|
| 22 |
|
| 23 |
# C2S-Scale-Gemma-2B model card
|
|
|
|
| 59 |
|
| 60 |
* Versatility: Demonstrates strong performance across a diverse set of single-cell and multi-cell tasks.
|
| 61 |
* Scalability: Trained on a massive dataset of over 57 million cells, showcasing the power of scaling LLMs for biological data.
|
| 62 |
+
* Generative Power: Capable of generating realistic single-cell gene expression profiles.
|
| 63 |
* Foundation for Fine-tuning: Can serve as a powerful pretrained foundation for specialized, domain-specific single-cell analysis tasks.
|
| 64 |
|
| 65 |
**Potential Applications**
|
|
|
|
| 121 |
).to(device)
|
| 122 |
|
| 123 |
# Format prompt (see previous section)
|
| 124 |
+
cell_sentence = "MALAT1 TMSB4X B2M EEF1A1 H3F3B ACTB FTL RPL13 ..." # Truncated for example, use at least 200 genes for inference
|
| 125 |
num_genes = 1000
|
| 126 |
organism = "Homo sapiens"
|
| 127 |
|
|
|
|
| 191 |
|
| 192 |
## License
|
| 193 |
|
| 194 |
+
The model weights shared on Huggingface are CC-by-4.0.
|
|
|
|
| 195 |
|
| 196 |
## Implementation information
|
| 197 |
|
|
|
|
| 248 |
# Gemma-2 Links
|
| 249 |
- HuggingFace: https://huggingface.co/google/gemma-2-2b
|
| 250 |
- Gemma-2 Blog Post: [Gemma explained: What's new in Gemma 2](https://developers.googleblog.com/en/gemma-explained-new-in-gemma-2/)
|
| 251 |
+
- Technical report: https://storage.googleapis.com/deepmind-media/gemma/gemma-2-report.pdf
|