README.md

Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


llama2_tifa_question_generation - GGUF
- Model creator: https://huggingface.co./tifa-benchmark/
- Original model: https://huggingface.co./tifa-benchmark/llama2_tifa_question_generation/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [llama2_tifa_question_generation.Q2_K.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q2_K.gguf) | Q2_K | 2.36GB |
| [llama2_tifa_question_generation.IQ3_XS.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [llama2_tifa_question_generation.IQ3_S.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [llama2_tifa_question_generation.Q3_K_S.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [llama2_tifa_question_generation.IQ3_M.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [llama2_tifa_question_generation.Q3_K.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q3_K.gguf) | Q3_K | 3.07GB |
| [llama2_tifa_question_generation.Q3_K_M.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [llama2_tifa_question_generation.Q3_K_L.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [llama2_tifa_question_generation.IQ4_XS.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [llama2_tifa_question_generation.Q4_0.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q4_0.gguf) | Q4_0 | 3.56GB |
| [llama2_tifa_question_generation.IQ4_NL.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [llama2_tifa_question_generation.Q4_K_S.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [llama2_tifa_question_generation.Q4_K.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q4_K.gguf) | Q4_K | 3.8GB |
| [llama2_tifa_question_generation.Q4_K_M.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [llama2_tifa_question_generation.Q4_1.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q4_1.gguf) | Q4_1 | 3.95GB |
| [llama2_tifa_question_generation.Q5_0.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q5_0.gguf) | Q5_0 | 4.33GB |
| [llama2_tifa_question_generation.Q5_K_S.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [llama2_tifa_question_generation.Q5_K.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q5_K.gguf) | Q5_K | 4.45GB |
| [llama2_tifa_question_generation.Q5_K_M.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [llama2_tifa_question_generation.Q5_1.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q5_1.gguf) | Q5_1 | 4.72GB |
| [llama2_tifa_question_generation.Q6_K.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q6_K.gguf) | Q6_K | 5.15GB |
| [llama2_tifa_question_generation.Q8_0.gguf](https://huggingface.co./RichardErkhov/tifa-benchmark_-_llama2_tifa_question_generation-gguf/blob/main/llama2_tifa_question_generation.Q8_0.gguf) | Q8_0 | 6.67GB |




Original model description:
---
license: apache-2.0
inference: true
widget:
 - text: "<s>[INST] <<SYS>>\nGiven an image description, generate one or two multiple-choice questions that verifies if the image description is correct.\nClassify each concept into a type (object, human, animal, food, activity, attribute, counting, color, material, spatial, location, shape, other), and then generate a question for each type.\n\n<</SYS>>\n\nDescription: a blue rabbit and a red plane [/INST] Entities:"
pipeline_tag: text-generation
tags:
- text-generation-inference
- llama2
- text-to-image
datasets:
- TIFA
language:
- en
---
Project page: <https://tifa-benchmark.github.io/>

This is the text parsing and question generation model for the ICCV 2023 paper [TIFA: Accurate and Interpretable Text-to-Image Faithfulness Evaluation with Question Answering](https://arxiv.org/abs/2303.11897)

We introduce TIFA (Text-to-Image Faithfulness evaluation with question Answering), an automatic evaluation metric that measures the faithfulness of a generated image to its text input via visual question answering (VQA). Specifically, given a text input, we automatically generate several question-answer pairs using a language model. We calculate image faithfulness by checking whether existing VQA models can answer these questions using the generated image. 

Specifically, this fine-tuned LLaMA 2 model is the substitute for the GPT-3 model in the paper. It can parse an arbitrary prompt into visual entities, attributes, relations, etc. and generate question-answer tuples for each of them. See examples below.


# QuickStart

All codes are from <https://github.com/Yushi-Hu/tifa>. Clone this repo to easily use this model together with other modules (e.g. VQA) provided in TIFA.

Please follow the prompt format, which will give the best performance.


```python
import torch
import transformers

# prepare the LLaMA 2 model
model_name = "tifa-benchmark/llama2_tifa_question_generation"
pipeline = transformers.pipeline(
    "text-generation",
    model=model_name,
    torch_dtype=torch.float16,
    device_map="auto",
)


# formating prompt following LLaMA 2 style
def create_qg_prompt(caption):
    INTRO_BLURB = "Given an image description, generate one or two multiple-choice questions that verifies if the image description is correct.\nClassify each concept into a type (object, human, animal, food, activity, attribute, counting, color, material, spatial, location, shape, other), and then generate a question for each type.\n"
    formated_prompt = f"<s>[INST] <<SYS>>\n{INTRO_BLURB}\n<</SYS>>\n\n"
    formated_prompt += f"Description: {caption} [/INST] Entities:"
    return formated_prompt


test_caption = "a blue rabbit and a red plane"

# create prompt
prompt = create_qg_prompt(text_caption)

# text completion
sequences = pipeline(
        prompt, do_sample=False, num_beams=5, num_return_sequences=1, max_length=512)
output = sequences[0]['generated_text'][len(prompt):]
output = output.split('\n\n')[0]

# output
print(output)

#### Expected output ###
#  rabbit, plane
# Activites:
# Colors: blue, red
# Counting:
# Other attributes:
# About rabbit (animal):
# Q: is this a rabbit?
# Choices: yes, no
# A: yes
# About rabbit (animal):
# Q: what animal is in the picture?
# Choices: rabbit, dog, cat, fish
# A: rabbit
# About plane (object):
# Q: is this a plane?
# Choices: yes, no
# A: yes
# About plane (object):
# Q: what type of vehicle is this?
# Choices: plane, car, motorcycle, bus
# A: plane
# About blue (color):
# Q: is the rabbit blue?
# Choices: yes, no
# A: yes
# About blue (color):
# Q: what color is the rabbit?
# Choices: blue, red, yellow, green
# A: blue
# About red (color):
# Q: is the plane red?
# Choices: yes, no
# A: yes
# About red (color):
# Q: what color is the plane?
# Choices: red, blue, yellow, green
# A: red
```

# Use this LM under tifascore package

tifascore provides extra functions to parse this output etc. First install tifascore according to <https://github.com/Yushi-Hu/tifa>. Then the usage is below

```python
from tifascore import get_llama2_pipeline, get_llama2_question_and_answers

pipeline = get_llama2_pipeline("tifa-benchmark/llama2_tifa_question_generation")

print(get_llama2_question_and_answers(pipeline, "a blue rabbit and a red plane"))

#### Expected output ###
# [{'caption': 'a blue rabbit and a red plane', 'element': 'rabbit', 'question': 'what animal is in the picture?', 'choices': ['rabbit', 'dog', 'cat', 'fish'], 'answer': 'rabbit', 'element_type': 'animal/human'}, {'caption': 'a blue rabbit and a red plane', 'element': 'plane', 'question': 'is this a plane?', 'choices': ['yes', 'no'], 'answer': 'yes', 'element_type': 'object'}, {'caption': 'a blue rabbit and a red plane', 'element': 'plane', 'question': 'what type of vehicle is this?', 'choices': ['plane', 'car', 'motorcycle', 'bus'], 'answer': 'plane', 'element_type': 'object'}, {'caption': 'a blue rabbit and a red plane', 'element': 'blue', 'question': 'is the rabbit blue?', 'choices': ['yes', 'no'], 'answer': 'yes', 'element_type': 'color'}, {'caption': 'a blue rabbit and a red plane', 'element': 'blue', 'question': 'what color is the rabbit?', 'choices': ['blue', 'red', 'yellow', 'green'], 'answer': 'blue', 'element_type': 'color'}, {'caption': 'a blue rabbit and a red plane', 'element': 'red', 'question': 'is the plane red?', 'choices': ['yes', 'no'], 'answer': 'yes', 'element_type': 'color'}, {'caption': 'a blue rabbit and a red plane', 'element': 'red', 'question': 'what color is the plane?', 'choices': ['red', 'blue', 'yellow', 'green'], 'answer': 'red', 'element_type': 'color'}]
```

## Bibtex
```
@article{hu2023tifa,
  title={Tifa: Accurate and interpretable text-to-image faithfulness evaluation with question answering},
  author={Hu, Yushi and Liu, Benlin and Kasai, Jungo and Wang, Yizhong and Ostendorf, Mari and Krishna, Ranjay and Smith, Noah A},
  journal={arXiv preprint arXiv:2303.11897},
  year={2023}
}
```