---
library_name: keras-hub
---
### Model Overview
An OPT decoder network.

This class implements a Transformer-based decoder model as described in
["OPT: Open Pre-trained Transformer Language Models"](https://arxiv.org/abs/2205.01068).
The default constructor gives a fully customizable, randomly initialized OPT
model with any number of layers, heads, and embedding dimensions. To load
preset architectures and weights, use the `from_preset()` constructor.

Disclaimer: Pre-trained models are provided on an "as is" basis, without
warranties or conditions of any kind. The underlying model is provided by a
third party and subject to a separate license, available
[here](https://github.com/facebookresearch/fairseq/).


__Arguments__


- __vocabulary_size__: int. The size of the token vocabulary.
- __num_layers__: int. The number of transformer decoder layers.
- __num_heads__: int. The number of attention heads for each transformer.
    The hidden size must be divisible by the number of attention heads.
- __hidden_dim__: int. The hidden size of the transformer decoder layers.
- __intermediate_dim__: int. The output dimension of the first Dense layer in
    a two-layer feedforward network for each transformer decoder layer.
- __dropout__: float. Dropout probability for the Transformer decoder.
- __max_sequence_length__: int. The maximum sequence length that this decoder
    can consume. If `None`, `max_sequence_length` uses the value from
    sequence length. This determines the variable shape for positional
    embeddings.

### Example Usage
```python
import keras
import keras_hub
import numpy as np
```

Use `generate()` to do text generation.
```python
opt_lm = keras_hub.models.OPTCausalLM.from_preset("opt_2.7b_en")
opt_lm.generate("I want to say", max_length=30)

# Generate with batched prompts.
opt_lm.generate(["This is a", "Where are you"], max_length=30)
```

Compile the `generate()` function with a custom sampler.
```python
opt_lm = keras_hub.models.OPTCausalLM.from_preset("opt_2.7b_en")
opt_lm.compile(sampler="greedy")
opt_lm.generate("I want to say", max_length=30)

opt_lm.compile(sampler=keras_hub.samplers.BeamSampler(num_beams=2))
opt_lm.generate("I want to say", max_length=30)
```

Use `generate()` without preprocessing.
```python
# Prompt the model with `5338, 318` (the token ids for `"Who is"`).
# Use `"padding_mask"` to indicate values that should not be overridden.
prompt = {
    "token_ids": np.array([[5338, 318, 0, 0, 0]] * 2),
    "padding_mask": np.array([[1, 1, 0, 0, 0]] * 2),
}

opt_lm = keras_hub.models.OPTCausalLM.from_preset(
    "opt_2.7b_en",
    preprocessor=None,
)
opt_lm.generate(prompt)
```

Call `fit()` on a single batch.
```python
features = ["The quick brown fox jumped.", "I forgot my homework."]
opt_lm = keras_hub.models.OPTCausalLM.from_preset("opt_2.7b_en")
opt_lm.fit(x=features, batch_size=2)
```

Call `fit()` without preprocessing.
```python
x = {
    "token_ids": np.array([[1, 2, 3, 4, 5]] * 2),
    "padding_mask": np.array([[1, 1, 1, 1, 1]] * 2),
}
y = np.array([[2, 3, 4, 5, 0]] * 2)
sw = np.array([[1, 1, 1, 1, 1]] * 2)

opt_lm = keras_hub.models.OPTCausalLM.from_preset(
    "opt_2.7b_en",
    preprocessor=None,
)
opt_lm.fit(x=x, y=y, sample_weight=sw, batch_size=2)
```

## Example Usage with Hugging Face URI

```python
import keras
import keras_hub
import numpy as np
```

Use `generate()` to do text generation.
```python
opt_lm = keras_hub.models.OPTCausalLM.from_preset("hf://keras/opt_2.7b_en")
opt_lm.generate("I want to say", max_length=30)

# Generate with batched prompts.
opt_lm.generate(["This is a", "Where are you"], max_length=30)
```

Compile the `generate()` function with a custom sampler.
```python
opt_lm = keras_hub.models.OPTCausalLM.from_preset("hf://keras/opt_2.7b_en")
opt_lm.compile(sampler="greedy")
opt_lm.generate("I want to say", max_length=30)

opt_lm.compile(sampler=keras_hub.samplers.BeamSampler(num_beams=2))
opt_lm.generate("I want to say", max_length=30)
```

Use `generate()` without preprocessing.
```python
# Prompt the model with `5338, 318` (the token ids for `"Who is"`).
# Use `"padding_mask"` to indicate values that should not be overridden.
prompt = {
    "token_ids": np.array([[5338, 318, 0, 0, 0]] * 2),
    "padding_mask": np.array([[1, 1, 0, 0, 0]] * 2),
}

opt_lm = keras_hub.models.OPTCausalLM.from_preset(
    "hf://keras/opt_2.7b_en",
    preprocessor=None,
)
opt_lm.generate(prompt)
```

Call `fit()` on a single batch.
```python
features = ["The quick brown fox jumped.", "I forgot my homework."]
opt_lm = keras_hub.models.OPTCausalLM.from_preset("hf://keras/opt_2.7b_en")
opt_lm.fit(x=features, batch_size=2)
```

Call `fit()` without preprocessing.
```python
x = {
    "token_ids": np.array([[1, 2, 3, 4, 5]] * 2),
    "padding_mask": np.array([[1, 1, 1, 1, 1]] * 2),
}
y = np.array([[2, 3, 4, 5, 0]] * 2)
sw = np.array([[1, 1, 1, 1, 1]] * 2)

opt_lm = keras_hub.models.OPTCausalLM.from_preset(
    "hf://keras/opt_2.7b_en",
    preprocessor=None,
)
opt_lm.fit(x=x, y=y, sample_weight=sw, batch_size=2)
```