# Copyright 2023 The EASYDEL Author @erfanzar (Erfan Zare Chavoshi).
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import typing as tp
from functools import partial
import chex
import jax
import jax.numpy as jnp
from flax import nnx as nn
from easydel.infra.base_module import EasyDeLBaseModule
from easydel.infra.factory import TaskType, register_module
from easydel.infra.modeling_outputs import (
BaseModelOutput,
CausalLMOutput,
SequenceClassifierOutput,
)
from easydel.infra.utils import (
auto_remat,
block_wise_ffn,
control_mlp_sharding,
get_dot_general_by_bits,
)
from easydel.layers.attention import AttentionModule, FlexibleAttentionModule
from easydel.layers.caching import (
PagedAttentionCache,
PagedAttentionCacheView,
PagedAttentionMetadata,
TransformerCache,
TransformerCacheView,
TransformerMetadata,
)
from easydel.layers.linear import ParallelLinear
from .cohere2_configuration import Cohere2Config
[docs]class Cohere2LayerNorm(nn.Module):
"""Cohere Layer Normalization.
Attributes:
dim (Union[int, tuple]): The dimension(s) to normalize over.
eps (float): A small epsilon value to prevent division by zero.
dtype (jnp.dtype): The data type for computation.
param_dtype (jnp.dtype): The data type for the parameters.
rngs (Optional[nn.Rngs]): Random number generators.
"""
def __init__(
self,
dim: tp.Union[int, tuple],
eps: float = 1e-6,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
rngs: nn.Rngs = None,
):
super().__init__()
if rngs is None:
rngs = nn.Rngs(0)
self.dim = dim
self.eps = eps
self.dtype = dtype
self.param_dtype = param_dtype
self.kernel = nn.Param(
nn.initializers.ones(
key=rngs.params(),
shape=(self.dim,) if isinstance(self.dim, int) else self.dim,
dtype=self.param_dtype,
),
)
def _norm(self, x: jnp.ndarray) -> jnp.ndarray:
"""Computes the Layer Normalization for a given input tensor."""
mean = jnp.mean(x, -1, keepdims=True)
variance = jnp.mean(jnp.pow((x - mean), 2), -1, keepdims=True)
return (x - mean) * jax.lax.rsqrt(variance + self.eps)
def __call__(self, x: jnp.ndarray) -> jnp.ndarray:
"""Applies Layer Normalization to the input tensor.
Args:
x (jnp.ndarray): The input tensor.
Returns:
jnp.ndarray: The normalized output tensor.
"""
if self.dtype in [
jnp.float8_e4m3b11fnuz,
jnp.float8_e4m3fn,
jnp.float8_e4m3fnuz,
jnp.float8_e5m2,
jnp.float8_e5m2fnuz,
]:
x = x.astype(jnp.float32)
else:
x = x.astype(jnp.promote_types(self.dtype, jnp.float32))
output = self._norm(x).astype(self.dtype)
weight = self.kernel.value.astype(self.dtype)
return output * weight
[docs]class Cohere2Attention(AttentionModule):
"""
Cohere2 Attention module, incorporating features like RoPE and sliding window attention.
Attributes:
config (Cohere2Config): Configuration object.
layer_idx (int): The index of the current layer.
dtype (jnp.dtype): Data type for computation.
param_dtype (jnp.dtype): Data type for parameters.
precision (jax.lax.PrecisionLike): JAX precision level.
rngs (nn.Rngs): Random number generators.
"""
def __init__(
self,
config: Cohere2Config,
layer_idx: int,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
) -> None:
super().__init__(config=config)
self.dtype = dtype
self.param_dtype = param_dtype
self.precision = precision
self.rngs = rngs
self.hidden_size = config.hidden_size
self.head_dim = self.config.hidden_size // self.config.num_attention_heads
self.num_key_value_groups = (
self.config.num_attention_heads // self.config.num_key_value_heads
)
if self.num_key_value_groups == 1:
assert self.config.num_attention_heads == self.config.num_key_value_heads
linear_class = partial(
ParallelLinear,
dtype=dtype,
param_dtype=param_dtype,
use_bias=config.attention_bias,
kernel_init=jax.nn.initializers.normal(config.initializer_range),
precision=self.precision,
rngs=rngs,
**get_dot_general_by_bits(config.bits, config.easy_method),
)
self.q_proj = linear_class(
config.hidden_size, config.num_attention_heads * self.head_dim
)
self.k_proj = linear_class(
config.hidden_size, config.num_key_value_heads * self.head_dim
)
self.v_proj = linear_class(
config.hidden_size, config.num_key_value_heads * self.head_dim
)
self.o_proj = linear_class(
config.num_attention_heads * self.head_dim, config.hidden_size
)
self.layer_idx = layer_idx
self.sliding_window = (
config.sliding_window
if (layer_idx + 1) % self.config.sliding_window_pattern != 0
else None
)
self.rotary = self.config.get_basic_rope(
self.dtype,
self.head_dim,
self.head_dim,
False,
)
self.attention_performer = FlexibleAttentionModule(
dropout_prob=config.attention_dropout,
base_config=config,
softmax_scale=self.head_dim**-0.5,
)
def __call__(
self,
hidden_states: chex.Array,
attention_mask: chex.Array,
position_ids: chex.Array,
causal_mask: tp.Optional[chex.Array | bool],
cache_view: tp.Optional[TransformerCacheView | PagedAttentionCacheView] = None,
cache_metadata: tp.Optional[TransformerMetadata | PagedAttentionMetadata] = None,
segment_ids: tp.Optional[chex.Array] = None,
output_attentions: bool = False,
fcm_mask: tp.Optional[chex.Array] = None,
frequencies: tp.Optional[chex.Array] = None,
):
"""Forward pass for the Cohere2 attention module.
Args:
hidden_states (chex.Array): Input hidden states.
attention_mask (chex.Array): Attention mask.
position_ids (chex.Array): Position IDs for RoPE.
causal_mask (Optional[chex.Array | bool]): Causal mask.
cache_view (Optional[TransformerCacheView | PagedAttentionCacheView]): Cache view for kv-caching.
cache_metadata (Optional[TransformerMetadata | PagedAttentionMetadata]): Metadata for paged attention.
segment_ids (Optional[chex.Array]): Segment IDs (if applicable).
output_attentions (bool): Whether to output attention weights.
fcm_mask (Optional[chex.Array]): FCM mask (if applicable).
frequencies (Optional[chex.Array]): Precomputed RoPE frequencies.
Returns:
Tuple[chex.Array, Optional[chex.Array]]: Attention output and optionally attention weights.
"""
batch_size, sequence_length = hidden_states.shape[:2]
(query_states, key_states, value_states) = (
self.q_proj(hidden_states),
self.k_proj(hidden_states),
self.v_proj(hidden_states),
)
query_states = query_states.reshape(
batch_size,
sequence_length,
self.config.num_attention_heads,
self.head_dim,
)
key_states = key_states.reshape(
batch_size,
sequence_length,
self.config.num_key_value_heads,
self.head_dim,
)
value_states = value_states.reshape(
batch_size,
sequence_length,
self.config.num_key_value_heads,
self.head_dim,
)
if self.sliding_window is not None:
query_states, key_states = self.rotary(
query=query_states,
key=key_states,
positions=position_ids,
frequencies=frequencies,
)
(
key_states,
value_states,
attention_mask,
init_attention_bias,
) = self.concatenate(
query=query_states,
key=key_states,
cache_view=cache_view,
value=value_states,
attention_mask=attention_mask,
causal_mask=causal_mask,
fcm_mask=fcm_mask,
sliding_windows=self.sliding_window,
)
attentions = self.attention_performer.forward(
query_states=query_states,
key_states=key_states,
value_states=value_states,
bias=None,
cache_metadata=cache_metadata,
cache_view=cache_view,
init_bias=init_attention_bias,
attention_mask=attention_mask,
segment_ids=segment_ids,
causal=True,
dropout_rng=self.rngs.params(),
)
attn_output = self.shard_attention_prod(
self._merge_heads(attentions.attention_outputs)
)
attn_output = self.o_proj(attn_output)
return attn_output, attentions.attention_weights
[docs]class Cohere2MLP(nn.Module):
def __init__(
self,
config: Cohere2Config,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
):
self.config = config
self.dtype = dtype
self.param_dtype = param_dtype
self.precision = precision
linear_class = partial(
ParallelLinear,
dtype=dtype,
param_dtype=param_dtype,
use_bias=False,
kernel_init=jax.nn.initializers.normal(config.initializer_range),
precision=self.precision,
rngs=rngs,
**get_dot_general_by_bits(config.bits, config.easy_method),
)
self.gate_proj = linear_class(config.hidden_size, config.intermediate_size)
self.down_proj = linear_class(config.intermediate_size, config.hidden_size)
self.up_proj = linear_class(config.hidden_size, config.intermediate_size)
def __call__(self, hidden_states: jnp.ndarray) -> jnp.ndarray:
hidden_states = control_mlp_sharding(hidden_states, self.config.partition_axis)
hidden_states = self.down_proj(
jax.nn.silu(self.gate_proj(hidden_states)) * self.up_proj(hidden_states)
)
return hidden_states
[docs]class Cohere2Block(nn.Module):
def __init__(
self,
config: Cohere2Config,
layer_idx: int,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
) -> None:
super().__init__()
self.config = config
self.layer_idx = layer_idx
self.dtype = dtype
self.param_dtype = param_dtype
self.precision = precision
self.rngs = rngs
attn_block = Cohere2Attention
mlp_block = Cohere2MLP
attn_block, mlp_block = auto_remat(
attn_block,
mlp_block,
policy=config.gradient_checkpointing,
)
self.self_attn = attn_block(
config,
layer_idx=layer_idx,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.mlp = mlp_block(
config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.input_layernorm = Cohere2LayerNorm(
self.config.hidden_size,
eps=self.config.layer_norm_eps,
dtype=dtype,
param_dtype=param_dtype,
rngs=rngs,
)
self.is_sliding = (layer_idx + 1) % self.config.sliding_window_pattern != 0
self.sliding_window = config.sliding_window
def __call__(
self,
hidden_states: chex.Array,
attention_mask: chex.Array,
position_ids: chex.Array,
causal_mask: tp.Optional[chex.Array | bool],
cache_view: tp.Optional[TransformerCacheView | PagedAttentionCacheView] = None,
cache_metadata: tp.Optional[TransformerMetadata | PagedAttentionMetadata] = None,
segment_ids: tp.Optional[chex.Array] = None,
output_attentions: bool = False,
fcm_mask: tp.Optional[chex.Array] = None,
frequencies: tp.Optional[chex.Array] = None,
):
"""
Forward pass of the module block.
Args:
hidden_states (chex.Array): Input hidden states.
attention_mask (chex.Array): Mask to apply on the attention scores.
position_ids (chex.Array): Position indices for the tokens.
causal_mask (chex.Array): Causal mask for ensuring autoregressive behavior.
segment_ids (tp.Optional[chex.Array]): Segment IDs for segment-based attention (optional).
deterministic (bool): If True, disables dropout for deterministic behavior.
init_cache (bool): If True, initializes cache for caching keys and values.
output_attentions (bool): If True, outputs attention weights alongside the hidden states.
fcm_mask (tp.Optional[chex.Array]): fcm mask to be combined with attn mask and causal mask.
Returns:
tp.Tuple[chex.Array, chex.Array]: A tuple containing the attention output and the attention weights.
"""
residual = hidden_states
hidden_states = self.input_layernorm(hidden_states)
attn_outputs = self.self_attn(
hidden_states,
attention_mask,
position_ids,
causal_mask,
cache_view,
cache_metadata,
segment_ids,
output_attentions,
fcm_mask,
frequencies,
)
attn_output = attn_outputs[0]
feed_forward_input = hidden_states
if self.config.use_scan_mlp:
feed_forward_hidden_states = block_wise_ffn(
self.mlp,
feed_forward_input,
self.config.scan_mlp_chunk_size,
)
else:
feed_forward_hidden_states = self.mlp(feed_forward_input)
hidden_states = attn_output + feed_forward_hidden_states + residual
return (hidden_states,) + attn_outputs[1:]
[docs]@register_module(
TaskType.BASE_MODULE,
config=Cohere2Config,
model_type="cohere2",
)
class Cohere2Model(EasyDeLBaseModule):
def __init__(
self,
config: Cohere2Config,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
):
super().__init__(
config=config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.embed_tokens = nn.Embed(
config.vocab_size,
config.hidden_size,
embedding_init=nn.initializers.normal(stddev=config.initializer_range),
dtype=dtype,
param_dtype=param_dtype,
rngs=rngs,
)
self.layers = [
Cohere2Block(
config=config,
layer_idx=idx,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
for idx in range(config.num_hidden_layers)
]
self.norm = Cohere2LayerNorm(
self.config.hidden_size,
eps=self.config.layer_norm_eps,
dtype=dtype,
param_dtype=param_dtype,
)
def __call__(
self,
input_ids: tp.Optional[chex.Array] = None,
inputs_embeds: tp.Optional[chex.Array] = None,
attention_mask: tp.Optional[chex.Array] = None,
position_ids: tp.Optional[chex.Array] = None,
segment_ids: tp.Optional[chex.Array] = None,
output_attentions: tp.Optional[bool] = None,
output_hidden_states: tp.Optional[bool] = None,
past_key_values: tp.Optional[TransformerCache | PagedAttentionCache] = None,
cache_metadata: tp.Optional[TransformerMetadata | PagedAttentionMetadata] = None,
return_dict: bool = True,
) -> tp.Union[BaseModelOutput, tp.Tuple]:
if (input_ids is None) ^ (inputs_embeds is not None):
raise ValueError(
"You cannot specify both input_ids and inputs_embeds at the same time, and must specify either one"
)
if inputs_embeds is None:
inputs_embeds = self.embed_tokens(input_ids.astype("i4"))
batch_size, sequence_length, _ = inputs_embeds.shape
all_attentions = () if output_attentions else None
all_hidden_states = () if output_hidden_states else None
assert sequence_length <= self.config.max_position_embeddings, (
f"Maximum Position Embedding Reached ! (Excepted <= {self.config.max_position_embeddings} got {sequence_length})"
)
if attention_mask is None:
attention_mask = jnp.ones((batch_size, sequence_length), "b1")
else:
if attention_mask.dtype != jnp.bool:
attention_mask = jnp.astype(attention_mask == 1, "b1")
if position_ids is None:
position_ids = jnp.broadcast_to(
jnp.clip(jnp.cumsum(attention_mask, axis=-1) - 1, a_min=0),
(batch_size, sequence_length),
).astype(jnp.int32)
hidden_states = inputs_embeds
if past_key_values is None:
past_key_values = TransformerCache.init_empty(len(self.layers))
for idx, block in enumerate(self.layers):
if output_hidden_states:
all_hidden_states += (hidden_states,)
layer_outputs = block(
hidden_states=hidden_states,
attention_mask=attention_mask,
position_ids=position_ids,
cache_view=past_key_values.views[idx],
cache_metadata=cache_metadata,
causal_mask=self.causal_mask,
output_attentions=output_attentions,
segment_ids=segment_ids,
frequencies=self.frequencies,
)
hidden_states = layer_outputs[0]
if output_attentions:
all_attentions += (layer_outputs[1],)
hidden_states = self.norm(hidden_states)
if output_hidden_states:
all_hidden_states = hidden_states[1] + (hidden_states,)
outputs = (hidden_states, all_hidden_states, all_attentions, past_key_values)
if not return_dict:
return tuple(v for v in outputs if v is not None)
return BaseModelOutput(
last_hidden_state=hidden_states,
hidden_states=all_hidden_states,
attentions=all_attentions,
past_key_values=past_key_values,
)
[docs]@register_module(
TaskType.CAUSAL_LM,
config=Cohere2Config,
model_type="cohere2",
)
class Cohere2ForCausalLM(EasyDeLBaseModule):
def __init__(
self,
config: Cohere2Config,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
):
super().__init__(
config=config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.model = Cohere2Model(
config=config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.lm_head = ParallelLinear(
config.hidden_size,
config.vocab_size,
dtype=dtype,
param_dtype=param_dtype,
use_bias=False,
kernel_init=nn.initializers.normal(stddev=config.initializer_range),
precision=precision,
rngs=rngs,
**get_dot_general_by_bits(config.bits, config.easy_method),
)
self.logit_scale = self.config.logit_scale
def __call__(
self,
input_ids: tp.Optional[chex.Array] = None,
inputs_embeds: tp.Optional[chex.Array] = None,
attention_mask: tp.Optional[chex.Array] = None,
position_ids: tp.Optional[chex.Array] = None,
segment_ids: tp.Optional[chex.Array] = None,
output_attentions: tp.Optional[bool] = None,
output_hidden_states: tp.Optional[bool] = None,
past_key_values: tp.Optional[TransformerCache | PagedAttentionCache] = None,
cache_metadata: tp.Optional[TransformerMetadata | PagedAttentionMetadata] = None,
return_dict: bool = True,
) -> tp.Union[CausalLMOutput, tp.Tuple]:
"""
Forward pass through the Cohere module.
Args:
input_ids (chex.Array): Input tensor containing token IDs.
attention_mask (chex.Array): Mask for attention.
position_ids (chex.Array): Positional indices.
segment_ids (tp.Optional[chex.Array]): Segment IDs for different input parts.
inputs_embeds (tp.Optional[chex.Array]): Embedded input tensor.
output_attentions (tp.Optional[bool]): If True, output attention weights.
output_hidden_states (tp.Optional[bool]): If True, output hidden states.
init_cache (bool): If True, initialize cache for decoding.
deterministic (bool): If True, disable dropout.
return_dict (bool): If True, return a dictionary of outputs.
Returns:
CausalLMOutput | tp.Tuple: Model output, either as a named tuple or a standard tuple.
"""
outputs = self.model(
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
past_key_values=past_key_values,
cache_metadata=cache_metadata,
return_dict=return_dict,
inputs_embeds=inputs_embeds,
segment_ids=segment_ids,
)
hidden_states = outputs[0]
if self.config.tie_word_embeddings:
lm_logits = jax.lax.dot_general(
hidden_states,
self.model.embed_tokens.embedding.value.T,
(((hidden_states.ndim - 1), (0,)), ((), ())),
)
else:
lm_logits = self.lm_head(hidden_states)
lm_logits = lm_logits * self.logit_scale
if not return_dict:
return (lm_logits,) + outputs[1:]
return CausalLMOutput(
logits=lm_logits,
hidden_states=outputs.hidden_states,
attentions=outputs.attentions,
past_key_values=outputs.past_key_values,
)
[docs]@register_module(
TaskType.SEQUENCE_CLASSIFICATION,
config=Cohere2Config,
model_type="cohere2",
)
class Cohere2ForSequenceClassification(EasyDeLBaseModule):
def __init__(
self,
config: Cohere2Config,
dtype: jnp.dtype = jnp.float32,
param_dtype: jnp.dtype = jnp.float32,
precision: jax.lax.PrecisionLike = None,
*,
rngs: nn.Rngs,
):
super().__init__(
config=config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
self.model = Cohere2Model(
config=config,
dtype=dtype,
param_dtype=param_dtype,
precision=precision,
rngs=rngs,
)
assert hasattr(config, "num_labels"), (
"in order to use `SequenceClassification` Models in `EasyDeL` you first need to attach `num_labels` to model `config`"
)
self.score = ParallelLinear(
config.hidden_size,
config.num_labels,
dtype=dtype,
param_dtype=param_dtype,
use_bias=False,
kernel_init=jax.nn.initializers.normal(stddev=config.initializer_range),
precision=precision,
rngs=rngs,
)
def __call__(
self,
input_ids: tp.Optional[chex.Array] = None,
inputs_embeds: tp.Optional[chex.Array] = None,
attention_mask: tp.Optional[chex.Array] = None,
position_ids: tp.Optional[chex.Array] = None,
segment_ids: tp.Optional[chex.Array] = None,
output_attentions: tp.Optional[bool] = None,
output_hidden_states: tp.Optional[bool] = None,
past_key_values: tp.Optional[TransformerCache | PagedAttentionCache] = None,
cache_metadata: tp.Optional[TransformerMetadata | PagedAttentionMetadata] = None,
return_dict: bool = True,
) -> tp.Union[SequenceClassifierOutput, tp.Tuple]:
transformer_outputs = self.model(
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_values=past_key_values,
cache_metadata=cache_metadata,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
return_dict=return_dict,
inputs_embeds=inputs_embeds,
segment_ids=segment_ids,
)
hidden_states = transformer_outputs[0]
logits = self.score(hidden_states)
if input_ids is not None:
batch_size = input_ids.shape[0]
else:
batch_size = inputs_embeds.shape[0]
if self.config.pad_token_id is None and batch_size != 1:
raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
if self.config.pad_token_id is None:
sequence_lengths = -1
else:
if input_ids is not None:
sequence_lengths = (
jnp.argmax(jnp.equal(input_ids, self.config.pad_token_id).astype("i4"), -1)
- 1
)
sequence_lengths = sequence_lengths % input_ids.shape[-1]
else:
sequence_lengths = -1
pooled_logits = logits[jnp.arange(batch_size), sequence_lengths]
if not return_dict:
output = (pooled_logits,) + transformer_outputs[1:]
return output
return SequenceClassifierOutput(
logits=pooled_logits,
past_key_values=past_key_values,
hidden_states=transformer_outputs.hidden_states,
attentions=transformer_outputs.attentions,
)