---
language: en
tags:
- seq2seq
- t5
- text-generation
- recipe-generation
pipeline_tag: text2text-generation
widget:
  - text: "provolone cheese, bacon, bread, ginger"
  - text: "sugar, crunchy jif peanut butter, cornflakes"
  - text: "sweet butter, confectioners sugar, flaked coconut, condensed milk, nuts, vanilla, dipping chocolate"
  - text: "macaroni, butter, salt, bacon, milk, flour, pepper, cream corn"
  - text: "hamburger, sausage, onion, regular, american cheese, colby cheese"
  - text: "chicken breasts, onion, garlic, great northern beans, black beans, green chilies, broccoli, garlic oil, butter, cajun seasoning, salt, oregano, thyme, black pepper, basil, worcestershire sauce, chicken broth, sour cream, chardonnay wine"
  - text: "serrano peppers, garlic, celery, oregano, canola oil, vinegar, water, kosher salt, salt, black pepper"
---

![avatar](chef-transformer.png)

# Chef Transformer (T5) 
> This is part of the
[Flax/Jax Community Week](https://discuss.huggingface.co/t/recipe-generation-model/7475), organized by [HuggingFace](https://huggingface.co/) and TPU usage sponsored by Google.


## Team Members
- Mehrdad Farahani ([m3hrdadfi](https://huggingface.co/m3hrdadfi))
- Nicholas Broad ([nbroad](https://huggingface.co/nbroad))
- Deepak Pandian ([rays2pix](https://huggingface.co/rays2pix))
- Kartik Godawat ([dk-crazydiv](https://huggingface.co/dk-crazydiv))

## Dataset

[RecipeNLG: A Cooking Recipes Dataset for Semi-Structured Text Generation](https://recipenlg.cs.put.poznan.pl/). This dataset contains **2,231,142** cooking recipes (>2 millions) with size of **2.14 GB**. It's processed in more careful way.

### Example

```json
{
  "NER": [
    "oyster crackers",
    "salad dressing",
    "lemon pepper",
    "dill weed",
    "garlic powder",
    "salad oil"
  ],
  "directions": [
    "Combine salad dressing mix and oil.",
    "Add dill weed, garlic powder and lemon pepper.",
    "Pour over crackers; stir to coat.",
    "Place in warm oven.",
    "Use very low temperature for 15 to 20 minutes."
  ],
  "ingredients": [
    "12 to 16 oz. plain oyster crackers",
    "1 pkg. Hidden Valley Ranch salad dressing mix",
    "1/4 tsp. lemon pepper",
    "1/2 to 1 tsp. dill weed",
    "1/4 tsp. garlic powder",
    "3/4 to 1 c. salad oil"
  ],
  "link": "www.cookbooks.com/Recipe-Details.aspx?id=648947",
  "source": "Gathered",
  "title": "Hidden Valley Ranch Oyster Crackers"
}
```

## How To Use

### Installing requirements

```bash
pip install transformers
```

```python
from transformers import FlaxAutoModelForSeq2SeqLM
from transformers import AutoTokenizer

MODEL_NAME_OR_PATH = "flax-community/t5-recipe-generation"
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME_OR_PATH, use_fast=True)
model = FlaxAutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME_OR_PATH)

prefix = "items: "
# generation_kwargs = {
#     "max_length": 1024,
#     "min_length": 128,
#     "no_repeat_ngram_size": 3,
#     "do_sample": True,
#     "top_k": 60,
#     "top_p": 0.95
# }
generation_kwargs = {
    "max_length": 512,
    "min_length": 64,
    "no_repeat_ngram_size": 3,
    "early_stopping": True,
    "num_beams": 5,
    "length_penalty": 1.5,
}

special_tokens = tokenizer.all_special_tokens
tokens_map = {
    "<sep>": "--",
    "<section>": "\n"
}
def skip_special_tokens(text, special_tokens):
    for token in special_tokens:
        text = text.replace(token, '')

    return text

def target_postprocessing(texts, special_tokens):
    if not isinstance(texts, list):
        texts = [texts]
    
    new_texts = []
    for text in texts:
        text = skip_special_tokens(text, special_tokens)

        for k, v in tokens_map.items():
            text = text.replace(k, v)

        new_texts.append(text)

    return new_texts

def generation_function(texts):
    _inputs = texts if isinstance(texts, list) else [texts]
    inputs = [prefix + inp for inp in _inputs]
    inputs = tokenizer(
        inputs, 
        max_length=256, 
        padding="max_length", 
        truncation=True, 
        return_tensors='jax'
    )

    input_ids = inputs.input_ids
    attention_mask = inputs.attention_mask

    output_ids = model.generate(
        input_ids=input_ids, 
        attention_mask=attention_mask,
        **generation_kwargs
    )
    generated = output_ids.sequences
    generated_recipe = target_postprocessing(
        tokenizer.batch_decode(generated, skip_special_tokens=False),
        special_tokens
    )
    return generated_recipe
```

```python
items = [
    "macaroni, butter, salt, bacon, milk, flour, pepper, cream corn"
]
generated = generation_function(items)
for text in generated:
    sections = text.split("\n")
    for section in sections:
        section = section.strip()
        if section.startswith("title:"):
            section = section.replace("title:", "")
            headline = "TITLE"
        elif section.startswith("ingredients:"):
            section = section.replace("ingredients:", "")
            headline = "INGREDIENTS"
        elif section.startswith("directions:"):
            section = section.replace("directions:", "")
            headline = "DIRECTIONS"
        
        section_info = [f"  - {i+1}: {info.strip().capitalize()}" for i, info in enumerate(section.split("--"))]
        print(f"[{headline}]:")
        print("\n".join(section_info))

    print("-" * 130)
```

Output:
```text
[TITLE]:
  - 1: Macaroni and corn
[INGREDIENTS]:
  - 1: 2 c. macaroni
  - 2: 2 tbsp. butter
  - 3: 1 tsp. salt
  - 4: 4 slices bacon
  - 5: 2 c. milk
  - 6: 2 tbsp. flour
  - 7: 1/4 tsp. pepper
  - 8: 1 can cream corn
[DIRECTIONS]:
  - 1: Cook macaroni in boiling salted water until tender.
  - 2: Drain.
  - 3: Melt butter in saucepan.
  - 4: Blend in flour, salt and pepper.
  - 5: Add milk all at once.
  - 6: Cook and stir until thickened and bubbly.
  - 7: Stir in corn and bacon.
  - 8: Pour over macaroni and mix well.
----------------------------------------------------------------------------------------------------------------------------------
```

## Evaluation

The following tables summarize the scores obtained by the **Chef Transformer**. Those marked as (*) are the baseline models.

|      Model      |  BLEU |  WER  | COSIM | ROUGE-2 |
|:---------------:|:-----:|:-----:|:-----:|:-------:|
|   Recipe1M+ *   | 0.844 | 0.786 | 0.589 |    -    |
|   RecipeNLG *   | 0.866 | 0.751 | 0.666 |    -    |
| ChefTransformer | 0.203 | 0.709 | 0.714 |  0.290  |


## Copyright

Special thanks to those who provided these fantastic materials.
- [Anatomy](https://www.flaticon.com/free-icon)
- [Chef Hat](https://www.vecteezy.com/members/jellyfishwater)