File size: 1,408 Bytes
4ff17fd efbad1e 8046475 efbad1e 8030636 efbad1e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
# T5-base data to text model specialized for Finance NLG
__complete version__
----
## Usage (HuggingFace Transformers)
#### Call the model
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("yseop/FNP_T5_D2T_complete")
model = AutoModelForSeq2SeqLM.from_pretrained("yseop/FNP_T5_D2T_complete")
text = ["Group profit | valIs | € 115.7 million && € 115.7 million | dTime | in 2019"]
```
#### Choose a generation method
```python
input_ids = tokenizer.encode(": {}".format(text), return_tensors="pt")
p = 0.82
k = 90
outputs = model.generate(input_ids,
do_sample=True,
top_p=p,
top_k=k,
early_stopping=True)
print(tokenizer.decode(outputs[0]))
```
```python
input_ids = tokenizer.encode(": {}".format(text), return_tensors="pt")
outputs = model.generate(input_ids,
max_length=200,
num_beams=2, repetition_penalty=2.5,
top_k=50, top_p=0.98,
length_penalty=1.0,
early_stopping=True)
print(tokenizer.decode(outputs[0]))
```
**Created by:** [Yseop](https://www.yseop.com/) | Pioneer in Natural Language Generation (NLG) technology. Scaling human expertise through Natural Language Generation. |