Zhengping/conditional-probability-regression

Model Card for Model ID

Model Details

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: [More Information Needed]
• Funded by [optional]: [More Information Needed]
• Shared by [optional]: [More Information Needed]
• Model type: [More Information Needed]
• Language(s) (NLP): en
• License: mit
• Finetuned from model [optional]: Qwen/Qwen2.5-14B-Instruct

Model Sources [optional]

• Repository: [More Information Needed]
• Paper [optional]: [More Information Needed]
• Demo [optional]: [More Information Needed]

Uses

Direct Use

import transformers
import torch
from transformers.pipelines import PIPELINE_REGISTRY
from transformers import (
    pipeline,
    AutoModelForCausalLM,
    PreTrainedTokenizer
)
from typing import (
    Dict,
    Callable,
    Tuple,
    List,
)
from src.pipelines.level_to_score_pipeline import LevelToScorePipeline
from src.rank_dicts import SingleLabelRankDict
from src.chat_templates import UNLITemplate


model = transformers.AutoModelForCausalLM.from_pretrained(
    "Zhengping/conditional-probability-regression",
    torch_dtype="auto",
    attn_implementation="flash_attention_2",
)
tokenizer = transformers.AutoTokenizer.from_pretrained(
    "Zhengping/conditional-probability-regression",
)

rank_dict = SingleLabelRankDict.from_tokenizer(tokenizer)

PIPELINE_REGISTRY.register_pipeline(
    "level-to-score",
    pipeline_class=LevelToScorePipeline,
    pt_model=AutoModelForCausalLM
)

# This allows fine-grained labeling, the greedy decoding gives a coarse score,
# one can also attach their own level-to-score function to the pipeline, e.g. using UNLI
# label transformation to get it more binarized
def _level_to_score_func(
    logits: Tuple[torch.FloatTensor],
    tokenizer: PreTrainedTokenizer
) -> Tuple[List[float], List[float]]:
    """ """
    logits = logits[0]
    num_labels = len(rank_dict)
    considering_ids = tokenizer.convert_tokens_to_ids([f" <|label_level_{i}|>" for i in range(num_labels)])
    selective_logits = torch.index_select(logits, 1, torch.tensor(considering_ids, device=logits.device))
    step_size = 1 / num_labels
    expectation = torch.tensor([[i * step_size + 1 / 2 * step_size for i in range(num_labels)]], device=selective_logits.device)
    scores = torch.softmax(selective_logits, dim=-1) @ expectation.T
    scores = scores.squeeze(-1).tolist()
    return scores, selective_logits.tolist()

pipe = pipeline(
    "level-to-score",
    model=model,
    max_new_tokens=2,
    tokenizer=tokenizer,
    device=0,
    level_to_score_func=_level_to_score_func,
    torch_dtype=torch.bfloat16,
)

template = UNLITemplate()

premise = "Sam is sleeping."
hypothesis = "Sam is awake."

inputs = template.get_prompt_template(premise=premise, hypothesis=hypothesis) +\
    template.get_completion_template(is_completion=True)
    
result = pipe(inputs)
print(result)

See our code repo for the definition of the scoring pipeline and templates.

Downstream Use [optional]

[More Information Needed]

Out-of-Scope Use

[More Information Needed]

Bias, Risks, and Limitations

[More Information Needed]

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

Training Details

Training Data

[More Information Needed]

Training Procedure

Preprocessing [optional]

[More Information Needed]

Training Hyperparameters

• Training regime: [More Information Needed]

Speeds, Sizes, Times [optional]

[More Information Needed]

Evaluation

Testing Data, Factors & Metrics

Testing Data

[More Information Needed]

Factors

[More Information Needed]

Metrics

[More Information Needed]

Results

[More Information Needed]

Summary

Model Examination [optional]

[More Information Needed]

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: [More Information Needed]
• Hours used: [More Information Needed]
• Cloud Provider: [More Information Needed]
• Compute Region: [More Information Needed]
• Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]