---
library_name: keras-hub
---
## Model Overview
ALBERT encoder network.

This class implements a bi-directional Transformer-based encoder as
described in
["ALBERT: A Lite BERT for Self-supervised Learning of Language Representations"](https://arxiv.org/abs/1909.11942).
ALBERT is a more efficient variant of BERT, and uses parameter reduction
techniques such as cross-layer parameter sharing and factorized embedding
parameterization. This model class includes the embedding lookups and
transformer layers, but not the masked language model or sentence order
prediction heads.

The default constructor gives a fully customizable, randomly initialized
ALBERT encoder 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.


__Arguments__


- __vocabulary_size__: int. The size of the token vocabulary.
- __num_layers__: int, must be divisible by `num_groups`. The number of
    "virtual" layers, i.e., the total number of times the input sequence
    will be fed through the groups in one forward pass. The input will
    be routed to the correct group based on the layer index.
- __num_heads__: int. The number of attention heads for each transformer.
    The hidden size must be divisible by the number of attention heads.
- __embedding_dim__: int. The size of the embeddings.
- __hidden_dim__: int. The size of the transformer encoding and pooler layers.
- __intermediate_dim__: int. The output dimension of the first Dense layer in
    a two-layer feedforward network for each transformer.
- __num_groups__: int. Number of groups, with each group having
    `num_inner_repetitions` number of `TransformerEncoder` layers.
- __num_inner_repetitions__: int. Number of `TransformerEncoder` layers per
    group.
- __dropout__: float. Dropout probability for the Transformer encoder.
- __max_sequence_length__: int. The maximum sequence length that this encoder
    can consume. If None, `max_sequence_length` uses the value from
    sequence length. This determines the variable shape for positional
    embeddings.
- __num_segments__: int. The number of types that the 'segment_ids' input can
    take.

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

Raw string data.
```python
features = ["The quick brown fox jumped.", "I forgot my homework."]
labels = [0, 3]

# Pretrained classifier.
classifier = keras_hub.models.AlbertClassifier.from_preset(
    "albert_large_en_uncased",
    num_classes=4,
)
classifier.fit(x=features, y=labels, batch_size=2)
classifier.predict(x=features, batch_size=2)

# Re-compile (e.g., with a new learning rate).
classifier.compile(
    loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    optimizer=keras.optimizers.Adam(5e-5),
    jit_compile=True,
)
# Access backbone programmatically (e.g., to change `trainable`).
classifier.backbone.trainable = False
# Fit again.
classifier.fit(x=features, y=labels, batch_size=2)
```

Preprocessed integer data.
```python
features = {
    "token_ids": np.ones(shape=(2, 12), dtype="int32"),
    "segment_ids": np.array([[0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0]] * 2),
    "padding_mask": np.array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]] * 2),
}
labels = [0, 3]

# Pretrained classifier without preprocessing.
classifier = keras_hub.models.AlbertClassifier.from_preset(
    "albert_large_en_uncased",
    num_classes=4,
    preprocessor=None,
)
classifier.fit(x=features, y=labels, batch_size=2)
```

## Example Usage with Hugging Face URI

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

Raw string data.
```python
features = ["The quick brown fox jumped.", "I forgot my homework."]
labels = [0, 3]

# Pretrained classifier.
classifier = keras_hub.models.AlbertClassifier.from_preset(
    "hf://keras/albert_large_en_uncased",
    num_classes=4,
)
classifier.fit(x=features, y=labels, batch_size=2)
classifier.predict(x=features, batch_size=2)

# Re-compile (e.g., with a new learning rate).
classifier.compile(
    loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    optimizer=keras.optimizers.Adam(5e-5),
    jit_compile=True,
)
# Access backbone programmatically (e.g., to change `trainable`).
classifier.backbone.trainable = False
# Fit again.
classifier.fit(x=features, y=labels, batch_size=2)
```

Preprocessed integer data.
```python
features = {
    "token_ids": np.ones(shape=(2, 12), dtype="int32"),
    "segment_ids": np.array([[0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0]] * 2),
    "padding_mask": np.array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]] * 2),
}
labels = [0, 3]

# Pretrained classifier without preprocessing.
classifier = keras_hub.models.AlbertClassifier.from_preset(
    "hf://keras/albert_large_en_uncased",
    num_classes=4,
    preprocessor=None,
)
classifier.fit(x=features, y=labels, batch_size=2)
```