# Inconsistency in Model Output \[ Token Classification\]

**URL:** <https://discuss.huggingface.co/t/inconsistency-in-model-output-token-classification/36320>\
**Category:** 🤗Transformers\
**Created:** [April 12, 2023, 8:58am UTC](https://discuss.huggingface.co/t/inconsistency-in-model-output-token-classification/36320 "2023-04-12T08:58:44Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![pchhapolika](https://avatars.discourse-cdn.com/v4/letter/p/b9e5f3/32.png) [@pchhapolika](https://discuss.huggingface.co/u/pchhapolika)\
**Post date:** [April 12, 2023, 8:58am UTC](https://discuss.huggingface.co/t/inconsistency-in-model-output-token-classification/36320/1 "2023-04-12T08:58:44Z")

</div>

```auto
from transformers import AutoTokenizer, AutoModelForTokenClassification
import torch
from transformers import pipeline

```

```auto
tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
model = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")

```

```auto
nlp = pipeline("ner", model=model, tokenizer=tokenizer)
example = "My name is Wolfgang and I live in Berlin"

```

```auto
ner_results = nlp(example)
print(ner_results)

```

**Output 1:**

```auto
[{'entity': 'B-PER', 'score': 0.9990139, 'index': 4, 'word': 'Wolfgang', 'start': 11, 'end': 19}, 
{'entity': 'B-LOC', 'score': 0.999645, 'index': 9, 'word': 'Berlin', 'start': 34, 'end': 40}]

```

```auto
inputs = tokenizer.encode_plus(example, return_tensors="pt", add_special_tokens=True, max_length=512, padding="max_length", truncation=True)
input_ids = inputs["input_ids"]
attention_mask = inputs["attention_mask"]
# Feed the encoded segment into the model to obtain the predicted labels for each token
outputs = model(input_ids, attention_mask=attention_mask)
logits = outputs.logits
predicted_labels = torch.argmax(logits, dim=2)[0]

```

`[0, 0, 0, 0, 3, 0, 0, 0, 0, 0]`

```auto
label_list = ["O","B-MISC","I-MISC","B-PER","I-PER","B-ORG","I-ORG","B-LOC","I-LOC"]
final_label_names = [label_list[label] for label in predicted_labels]

```

**Output 2:**

`['O','O','O','O', 'B-PER', 'O','O','O','O','O']`