File size: 1,796 Bytes
080fcfd
 
689598d
 
 
080fcfd
 
689598d
080fcfd
 
 
689598d
 
080fcfd
 
689598d
080fcfd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
63
---
license: apache-2.0
language: en
library_name: transformers
pipeline_tag: text-classification
tags:
- text-classification
- distilbert
- log-analysis
- openstack
- fine-tuned
widget:
- text: "Instance 1234 has failed to connect to the network"
---


# INFRNCE BERT Log Classification Model

This is a fine-tuned DistilBERT model for classifying OpenStack Nova log entries into different operational categories.

## Model Details

- **Base Model**: distilbert-base-uncased
- **Task**: Multi-class text classification
- **Number of Labels**: 6
- **Domain**: OpenStack log analysis

## Labels

The model classifies logs into the following categories:

- Error_Handling, - Instance_Management, - Network_Operations, - Resource_Management, - Scheduler_Operations, - System_Operations

## Usage

```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load the model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("your-username/infrnce-bert-log-classifier")
model = AutoModelForSequenceClassification.from_pretrained("your-username/infrnce-bert-log-classifier")

# Example usage
log_text = "Your OpenStack log entry here"
inputs = tokenizer(log_text, return_tensors="pt", truncation=True, padding=True, max_length=512)

with torch.no_grad():
    outputs = model(**inputs)
    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predicted_class_id = predictions.argmax().item()

print(f"Predicted class: {model.config.id2label[predicted_class_id]}")
```

## Training Data

The model was trained on a curated dataset of OpenStack Nova logs with both regex-based classifications and semantic clustering.

## Performance

The model was trained with controlled accuracy to achieve optimal performance on log classification tasks.