Update the Phi model to use the updated architecture.

candle-transformers/src/models/mod.rs

@@ -17,6 +17,7 @@ pub mod mixformer;
 pub mod mixtral;
 pub mod mpt;
 pub mod persimmon;
+pub mod phi;
 pub mod quantized_blip;
 pub mod quantized_blip_text;
 pub mod quantized_llama;

candle-transformers/src/models/phi.rs

@@ -0,0 +1,212 @@
+#![allow(unused)]
+use crate::models::with_tracing::{layer_norm, linear, Embedding, LayerNorm, Linear};
+/// Phi model.
+/// https://huggingface.co/microsoft/phi-2
+/// There is an alternative implementation of the phi model in mixformers.rs.
+/// This corresponds to the model update made with the following commit:
+/// https://huggingface.co/microsoft/phi-2/commit/cb2f4533604d8b67de604e7df03bfe6f3ca22869
+use candle::{DType, Device, IndexOp, Module, Result, Tensor, D};
+use candle_nn::{Activation, VarBuilder};
+use serde::Deserialize;
+
+// https://huggingface.co/microsoft/phi-2/blob/main/configuration_phi.py
+#[derive(Debug, Clone, PartialEq, Deserialize)]
+pub struct Config {
+    pub(crate) vocab_size: usize,
+    pub(crate) hidden_size: usize,
+    pub(crate) intermediate_size: usize,
+    pub(crate) num_hidden_layers: usize,
+    pub(crate) num_attention_heads: usize,
+    pub(crate) num_key_value_heads: Option<usize>,
+    pub(crate) hidden_act: Activation,
+    pub(crate) max_position_embeddings: usize,
+    pub(crate) layer_norm_eps: f64,
+    pub(crate) tie_word_embeddings: bool,
+    pub(crate) rope_theta: f32,
+    pub(crate) partial_rotary_factor: f64,
+    pub(crate) qk_layer_norm: bool,
+}
+
+impl Config {
+    fn num_key_value_heads(&self) -> usize {
+        self.num_key_value_heads.unwrap_or(self.num_attention_heads)
+    }
+
+    fn head_dim(&self) -> usize {
+        self.hidden_size / self.num_attention_heads
+    }
+}
+
+#[derive(Debug, Clone)]
+struct RotaryEmbedding {
+    sin: Tensor,
+    cos: Tensor,
+}
+
+impl RotaryEmbedding {
+    fn new(cfg: &Config, dev: &Device) -> Result<Self> {
+        let dim = (cfg.partial_rotary_factor * cfg.head_dim() as f64) as usize;
+        let inv_freq: Vec<_> = (0..dim)
+            .step_by(2)
+            .map(|i| 1f32 / cfg.rope_theta.powf(i as f32 / dim as f32))
+            .collect();
+        let inv_freq_len = inv_freq.len();
+        let inv_freq = Tensor::from_vec(inv_freq, (1, inv_freq_len), dev)?;
+        let t = Tensor::arange(0u32, cfg.max_position_embeddings as u32, dev)?
+            .to_dtype(DType::F32)?
+            .reshape((cfg.max_position_embeddings, 1))?;
+        let freqs = t.matmul(&inv_freq)?;
+        Ok(Self {
+            sin: freqs.sin()?,
+            cos: freqs.cos()?,
+        })
+    }
+}
+
+#[derive(Debug, Clone)]
+#[allow(clippy::upper_case_acronyms)]
+struct MLP {
+    fc1: Linear,
+    fc2: Linear,
+    act: Activation,
+}
+
+impl MLP {
+    fn new(cfg: &Config, vb: VarBuilder) -> Result<Self> {
+        let fc1 = linear(cfg.hidden_size, cfg.intermediate_size, vb.pp("fc1"))?;
+        let fc2 = linear(cfg.intermediate_size, cfg.hidden_size, vb.pp("fc2"))?;
+        Ok(Self {
+            fc1,
+            fc2,
+            act: cfg.hidden_act,
+        })
+    }
+}
+
+impl Module for MLP {
+    fn forward(&self, xs: &Tensor) -> Result<Tensor> {
+        xs.apply(&self.fc1)?.apply(&self.act)?.apply(&self.fc2)
+    }
+}
+
+#[derive(Clone)]
+struct Attention {
+    q_proj: Linear,
+    k_proj: Linear,
+    v_proj: Linear,
+    dense: Linear,
+    q_layernorm: Option<LayerNorm>,
+    k_layernorm: Option<LayerNorm>,
+    rotary_emb: RotaryEmbedding,
+    softmax_scale: f64,
+    num_heads: usize,
+    num_kv_heads: usize,
+    span: tracing::Span,
+}
+
+fn masked_fill(on_false: &Tensor, mask: &Tensor, on_true: f32) -> Result<Tensor> {
+    let shape = mask.shape();
+    let on_true = Tensor::new(on_true, on_false.device())?.broadcast_as(shape.dims())?;
+    let m = mask.where_cond(&on_true, on_false)?;
+    Ok(m)
+}
+
+impl Attention {
+    fn new(cfg: &Config, vb: VarBuilder) -> Result<Self> {
+        let num_heads = cfg.num_attention_heads;
+        let num_kv_heads = cfg.num_key_value_heads();
+        let head_dim = cfg.head_dim();
+        let q_proj = linear(cfg.hidden_size, num_heads * head_dim, vb.pp("q_proj"))?;
+        let k_proj = linear(cfg.hidden_size, num_kv_heads * head_dim, vb.pp("k_proj"))?;
+        let v_proj = linear(cfg.hidden_size, num_kv_heads * head_dim, vb.pp("v_proj"))?;
+        let dense = linear(num_heads * head_dim, cfg.hidden_size, vb.pp("dense"))?;
+        // Alternative rope scalings are not supported.
+        let rotary_emb = RotaryEmbedding::new(cfg, vb.device())?;
+        let (q_layernorm, k_layernorm) = if cfg.qk_layer_norm {
+            let q_layernorm = layer_norm(head_dim, cfg.layer_norm_eps, vb.pp("q_layernorm"))?;
+            let k_layernorm = layer_norm(head_dim, cfg.layer_norm_eps, vb.pp("k_layernorm"))?;
+            (Some(q_layernorm), Some(k_layernorm))
+        } else {
+            (None, None)
+        };
+        let softmax_scale = 1f64 / (head_dim as f64).sqrt();
+        Ok(Self {
+            q_proj,
+            k_proj,
+            v_proj,
+            dense,
+            q_layernorm,
+            k_layernorm,
+            rotary_emb,
+            softmax_scale,
+            num_heads,
+            num_kv_heads,
+            span: tracing::span!(tracing::Level::TRACE, "attention"),
+        })
+    }
+
+    fn forward(&mut self, xs: &Tensor, mask: Option<&Tensor>) -> Result<Tensor> {
+        let _enter = self.span.enter();
+        let (b_size, seq_len, _n_embd) = xs.dims3()?;
+        let query_states = self.q_proj.forward(xs)?;
+        let key_states = self.k_proj.forward(xs)?;
+        let value_states = self.v_proj.forward(xs)?;
+
+        let query_states = match &self.q_layernorm {
+            None => query_states,
+            Some(ln) => query_states.apply(ln)?,
+        };
+        let key_states = match &self.k_layernorm {
+            None => key_states,
+            Some(ln) => key_states.apply(ln)?,
+        };
+
+        let query_states = query_states
+            .reshape((b_size, seq_len, self.num_heads))?
+            .transpose(1, 2)?;
+        let key_states = key_states
+            .reshape((b_size, seq_len, self.num_kv_heads))?
+            .transpose(1, 2)?;
+        let value_states = value_states
+            .reshape((b_size, seq_len, self.num_kv_heads))?
+            .transpose(1, 2)?;
+
+        // TODO: rotary embeddings.
+        // TODO: KV cache
+        // TODO: repeat kv.
+        let attn_weights = (query_states
+            .to_dtype(DType::F32)?
+            .matmul(&key_states.to_dtype(DType::F32)?.t()?)?
+            * self.softmax_scale)?;
+        let attn_weights = match mask {
+            None => attn_weights,
+            Some(mask) => masked_fill(
+                &attn_weights,
+                &mask.broadcast_left(b_size * self.num_heads)?,
+                f32::NEG_INFINITY,
+            )?,
+        };
+        let attn_weights =
+            candle_nn::ops::softmax_last_dim(&attn_weights)?.to_dtype(value_states.dtype())?;
+        let attn_output = attn_weights.matmul(&value_states)?;
+        let attn_output = attn_output
+            .transpose(1, 2)?
+            .reshape((b_size, seq_len, ()))?;
+        attn_output.apply(&self.dense)
+    }
+}
+
+#[derive(Clone)]
+struct DecoderLayer {
+    self_attn: Attention,
+    mlp: MLP,
+    input_layernorm: LayerNorm,
+}
+
+#[derive(Clone)]
+pub struct Model {
+    embed_tokens: Embedding,
+    layers: Vec<DecoderLayer>,
+    final_layernorm: LayerNorm,
+    lm_head: Linear,
+}