Instructions to use yarongef/DistilProtBert with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use yarongef/DistilProtBert with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="yarongef/DistilProtBert")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("yarongef/DistilProtBert") model = AutoModelForMaskedLM.from_pretrained("yarongef/DistilProtBert", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -16,10 +16,10 @@ In addition to cross entropy and cosine teacher-student losses, DistilProtBert w
|
|
| 16 |
|
| 17 |
DistilProtBert was pretrained on millions of proteins sequences.
|
| 18 |
|
| 19 |
-
|
| 20 |
1. Size of the model:
|
| 21 |
-
|
| 22 |
-
|
| 23 |
2. Size of the pretraining dataset: ~43M proteins (ProtBert was pretrained on 216M proteins)
|
| 24 |
3. Hardware used for pretraining: five v100 32GB Nvidia GPUs (ProtBert was pretrained on 512 16GB TPUs)
|
| 25 |
|
|
|
|
| 16 |
|
| 17 |
DistilProtBert was pretrained on millions of proteins sequences.
|
| 18 |
|
| 19 |
+
Differences between DistilProtBert model and ProtBert:
|
| 20 |
1. Size of the model:
|
| 21 |
+
- 230M parameters (420M parameters in ProtBert)
|
| 22 |
+
- 15 hidden layers (30 hidden layers in ProtBert)
|
| 23 |
2. Size of the pretraining dataset: ~43M proteins (ProtBert was pretrained on 216M proteins)
|
| 24 |
3. Hardware used for pretraining: five v100 32GB Nvidia GPUs (ProtBert was pretrained on 512 16GB TPUs)
|
| 25 |
|