BERT-Based Classification Model for Optimal Temperature Selection

Overview

• Causal Reasoning
• Creativity
• In-Context Learning
• Instruction Following
• Machine Translation
• Summarization

Features

• Pre-trained Model: Uses the multilingual BERT model: Volavion/bert-base-multilingual-uncased-Temperature-CLS.
• Tokenization: Processes text inputs into numerical formats compatible with the model.
• Classification Output: Provides probabilities for each class, allowing precise evaluation of the prompt's capabilities.

Installation

1. Clone the repository if necessary:
   bash

   1git clone https://huggingface.co/Volavion/bert-base-multilingual-uncased-temperature-cls 2cd bert-base-multilingual-uncased-temperature-cls
2. Install the required Python libraries:
   pip install transformers torch numpy

Usage

1. Load the tokenizer and model:
   python

   1from transformers import AutoTokenizer, AutoModelForSequenceClassification 2 3model_name = "Volavion/bert-base-multilingual-uncased-Temperature-CLS" 4tokenizer = AutoTokenizer.from_pretrained(model_name, do_lower_case=True) 5model = AutoModelForSequenceClassification.from_pretrained(model_name)
2. Tokenize your input text:
   python

   1input_text = "Your input prompt here." 2encoded_dict = tokenizer.encode_plus( 3 input_text, 4 add_special_tokens=True, 5 max_length=512, 6 pad_to_max_length=True, 7 return_attention_mask=True, 8 return_tensors="pt" 9)
3. Perform inference:
   python

   1import torch 2import numpy as np 3 4input_ids = encoded_dict["input_ids"].to(device) 5attention_mask = encoded_dict["attention_mask"].to(device) 6 7model.eval() 8with torch.no_grad(): 9 outputs = model(input_ids, attention_mask=attention_mask) 10 11logits = outputs.logits.cpu().numpy() 12probabilities = np.exp(logits - np.max(logits, axis=1, keepdims=True)) 13probabilities /= np.sum(probabilities, axis=1, keepdims=True)
4. Map probabilities to abilities:
   python

   1ability_mapping = {0: "Causal Reasoning", 1: "Creativity", 2: "In-Context Learning", 2 3: "Instruction Following", 4: "Machine Translation", 5: "Summarization"} 3for prob, ability in zip(probabilities[0], ability_mapping.values()): 4 print(f"{ability}: {prob*100:.2f}%")

Example Output

Device Compatibility

Contributing

License