Update README.md
Browse files
README.md
CHANGED
|
@@ -39,8 +39,7 @@ Segment-NT-multi-species has been shown to generalize up to sequences of 50,000
|
|
| 39 |
the `rescaling_factor` of the Rotary Embedding layer in the esm model `num_dna_tokens_inference / max_num_tokens_nt` where `num_dna_tokens_inference` is the number of tokens at inference
|
| 40 |
(i.e 6669 for a sequence of 40008 base pairs) and `max_num_tokens_nt` is the max number of tokens on which the backbone nucleotide-transformer was trained on, i.e `2048`.
|
| 41 |
|
| 42 |
-
|
| 43 |
-
The `./inference_segment_nt.ipynb` notebook shows how to set the rescaling factor and infer on a 50kb genic sequence of the human chromosome 20.
|
| 44 |
|
| 45 |
```python
|
| 46 |
# Load model and tokenizer
|
|
|
|
| 39 |
the `rescaling_factor` of the Rotary Embedding layer in the esm model `num_dna_tokens_inference / max_num_tokens_nt` where `num_dna_tokens_inference` is the number of tokens at inference
|
| 40 |
(i.e 6669 for a sequence of 40008 base pairs) and `max_num_tokens_nt` is the max number of tokens on which the backbone nucleotide-transformer was trained on, i.e `2048`.
|
| 41 |
|
| 42 |
+
The `./inference_segment_nt.ipynb` has been set up to run in Google Colab and shows how to set the rescaling factor and infer on a 50kb genic sequence of the human chromosome 20.
|
|
|
|
| 43 |
|
| 44 |
```python
|
| 45 |
# Load model and tokenizer
|