Token Classification
Transformers
PyTorch
English
T5
NER
token classification
information extraction
question answering
Instructions to use knowledgator/UTC-T5-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use knowledgator/UTC-T5-large with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="knowledgator/UTC-T5-large")# Load model directly from transformers import AutoModelForTokenClassification model = AutoModelForTokenClassification.from_pretrained("knowledgator/UTC-T5-large", device_map="auto") - Notebooks
- Google Colab
- Kaggle
fix t5 model initalization
Browse files
README.md
CHANGED
|
@@ -32,7 +32,10 @@ We recommend to use the model with transformers `ner` pipeline:
|
|
| 32 |
|
| 33 |
```python
|
| 34 |
from transformers import AutoTokenizer, T5PreTrainedModel, T5Config, T5EncoderModel
|
|
|
|
|
|
|
| 35 |
from transformers import pipeline
|
|
|
|
| 36 |
|
| 37 |
class T5EncoderForTokenClassification(T5PreTrainedModel):
|
| 38 |
_tied_weights_keys = ["encoder.embed_tokens.weight"]
|
|
@@ -155,7 +158,7 @@ def process(text, prompt, treshold=0.5):
|
|
| 155 |
|
| 156 |
return processed_results
|
| 157 |
|
| 158 |
-
tokenizer = AutoTokenizer.from_pretrained("knowledgator/UTC-
|
| 159 |
model = T5EncoderForTokenClassification.from_pretrained("knowledgator/UTC-T5-large")
|
| 160 |
|
| 161 |
nlp = pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy = 'first')
|
|
|
|
| 32 |
|
| 33 |
```python
|
| 34 |
from transformers import AutoTokenizer, T5PreTrainedModel, T5Config, T5EncoderModel
|
| 35 |
+
from transformers.modeling_outputs import TokenClassifierOutput
|
| 36 |
+
from typing import Union, Optional, Tuple
|
| 37 |
from transformers import pipeline
|
| 38 |
+
import torch
|
| 39 |
|
| 40 |
class T5EncoderForTokenClassification(T5PreTrainedModel):
|
| 41 |
_tied_weights_keys = ["encoder.embed_tokens.weight"]
|
|
|
|
| 158 |
|
| 159 |
return processed_results
|
| 160 |
|
| 161 |
+
tokenizer = AutoTokenizer.from_pretrained("knowledgator/UTC-T5-large")
|
| 162 |
model = T5EncoderForTokenClassification.from_pretrained("knowledgator/UTC-T5-large")
|
| 163 |
|
| 164 |
nlp = pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy = 'first')
|