updated to the newest version of burn

Cargo.toml

@@ -8,17 +8,13 @@ edition = "2021"
 [features]
 wgpu-backend = ["burn-wgpu"]
 
-[dependencies.burn-wgpu]
-package = "burn-wgpu"
-git = "https://github.com/burn-rs/burn.git"
-optional = true
-
 [dependencies]
-burn = "0.14.0"
+burn = "0.20.1"
-burn-ndarray = "0.14.0"
+burn-ndarray = "0.20.1"
-burn-tch = "0.14.0"
+burn-tch = "0.20.1"
-burn-autodiff = "0.14.0"
+burn-autodiff = "0.20.1"
-tch = "0.15.0"
+burn-wgpu = { version = "0.20.1", optional = true }
+tch = "0.22.0"
 serde = {version = "1.0.171", features = ["std", "derive"]}
 npy = "0.4.0"
 num-traits = "0.2.15"

src/bin/sample/main.rs

@@ -12,7 +12,7 @@ use burn::{
 
 cfg_if::cfg_if! {
     if #[cfg(feature = "wgpu-backend")] {
-        use burn_wgpu::{WgpuBackend, WgpuDevice, AutoGraphicsApi};
+        use burn_wgpu::{Wgpu, WgpuDevice};
     } else {
         use burn_tch::{LibTorch, LibTorchDevice};
     }
@@ -58,7 +58,7 @@ fn main() {
 
     cfg_if::cfg_if! {
         if #[cfg(feature = "wgpu-backend")] {
-            type Backend = WgpuBackend<AutoGraphicsApi, f32, i32>;
+            type Backend = Wgpu;
             let device = WgpuDevice::BestAvailable;
         } else {
             type Backend = LibTorch<f32>;

src/model/autoencoder/mod.rs

@@ -23,7 +23,7 @@ use crate::backend::{qkv_attention, attn_decoder_mask};
 
 use std::iter;
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct AutoencoderConfig {}
 
 impl AutoencoderConfig {
@@ -71,7 +71,7 @@ impl<B: Backend> Autoencoder<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct EncoderConfig {
     channels: Vec<(usize, usize)>,
     n_group: usize,
@@ -144,7 +144,7 @@ impl<B: Backend> Encoder<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct DecoderConfig {
     channels: Vec<(usize, usize)>,
     n_group: usize,
@@ -216,7 +216,7 @@ impl<B: Backend> Decoder<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct EncoderBlockConfig {
     n_channels_in: usize,
     n_channels_out: usize,
@@ -265,7 +265,7 @@ impl<B: Backend> EncoderBlock<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct DecoderBlockConfig {
     n_channels_in: usize,
     n_channels_out: usize,
@@ -323,7 +323,7 @@ impl<B: Backend> DecoderBlock<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct PaddedConv2dConfig {
     channels: [usize; 2],
     kernel_size: usize,
@@ -427,7 +427,7 @@ pub struct Padding {
     pad_bottom: usize,
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct MidConfig {
     n_channel: usize,
 }
@@ -462,7 +462,7 @@ impl<B: Backend> Mid<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResnetBlockConfig {
     in_channels: usize,
     out_channels: usize,
@@ -527,7 +527,7 @@ impl<B: Backend> ResnetBlock<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ConvSelfAttentionBlockConfig {
     n_channel: usize,
 }

src/model/clip/mod.rs

@@ -15,7 +15,7 @@ use burn::{
 //use crate::backend::Backend as MyBackend;
 use crate::backend::{qkv_attention, attn_decoder_mask};
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct CLIPConfig {
     n_vocab: usize,
     n_state: usize,
@@ -75,7 +75,7 @@ impl<B: Backend> CLIP<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResidualDecoderAttentionBlockConfig {
     n_state: usize,
     n_head: usize,
@@ -114,7 +114,7 @@ impl<B: Backend> ResidualDecoderAttentionBlock<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct MultiHeadSelfAttentionConfig {
     n_state: usize,
     n_head: usize,

src/model/groupnorm/mod.rs

@@ -6,7 +6,7 @@ use burn::{
     tensor::{backend::Backend, Tensor},
 };
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct GroupNormConfig {
     n_group: usize,
     n_channel: usize,

src/model/load.rs

@@ -9,7 +9,7 @@ use burn::{
     config::Config,
     module::{Module, Param},
     nn::{self, conv},
-    tensor::{backend::Backend, Data, Tensor},
+    tensor::{backend::Backend, Tensor},
 };
 
 use burn::tensor::ElementConversion;
@@ -98,7 +98,7 @@ pub fn load_layer_norm<B: Backend>(
 
     let mut layer_norm = nn::LayerNormConfig::new(n_state).with_epsilon(eps).init(device);
     layer_norm.gamma = Param::from_tensor(weight);
-    layer_norm.beta = Param::from_tensor(bias);
+    layer_norm.beta = Some(Param::from_tensor(bias));
 
     Ok(layer_norm)
 }

src/model/stablediffusion/mod.rs

@@ -3,7 +3,7 @@ pub mod load;
 use burn::{
     config::Config,
     module::{Module, Param},
-    tensor::{backend::Backend, BasicOps, Data, Distribution, Float, Int, Tensor},
+    tensor::{backend::Backend, BasicOps, Distribution, Float, Int, Tensor},
     tensor::cast::ToElement, 
 };
 
@@ -16,7 +16,7 @@ use super::clip::{CLIPConfig, CLIP};
 use super::unet::{UNet, UNetConfig};
 use crate::tokenizer::SimpleTokenizer;
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct StableDiffusionConfig {}
 
 impl StableDiffusionConfig {
@@ -192,7 +192,7 @@ impl<B: Backend> StableDiffusion<B> {
     }
 
     pub fn unconditional_context(&self, tokenizer: &SimpleTokenizer) -> Tensor<B, 2> {
-        self.context(tokenizer, "").squeeze(0)
+        self.context(tokenizer, "").squeeze::<2>()
     }
 
     pub fn context(&self, tokenizer: &SimpleTokenizer, text: &str) -> Tensor<B, 3> {

src/model/unet/mod.rs

@@ -29,7 +29,7 @@ fn timestep_embedding<B: Backend>(
     Tensor::cat(vec![args.clone().cos(), args.sin()], 0).unsqueeze()
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct UNetConfig {}
 
 impl UNetConfig {
@@ -196,7 +196,7 @@ trait UNetBlock<B: Backend> {
     fn forward(&self, x: Tensor<B, 4>, emb: Tensor<B, 2>, context: Tensor<B, 3>) -> Tensor<B, 4>;
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResTransformerConfig {
     n_channels_in: usize,
     n_channels_embed: usize,
@@ -235,7 +235,7 @@ impl<B: Backend> UNetBlock<B> for ResTransformer<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResUpSampleConfig {
     n_channels_in: usize,
     n_channels_embed: usize,
@@ -270,7 +270,7 @@ impl<B: Backend> UNetBlock<B> for ResUpSample<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResTransformerUpsampleConfig {
     n_channels_in: usize,
     n_channels_embed: usize,
@@ -316,7 +316,7 @@ impl<B: Backend> UNetBlock<B> for ResTransformerUpsample<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResTransformerResConfig {
     n_channels_in: usize,
     n_channels_embed: usize,
@@ -367,7 +367,7 @@ impl<B: Backend> UNetBlock<B> for ResTransformerRes<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct UpsampleConfig {
     n_channels: usize,
 }
@@ -404,7 +404,7 @@ impl<B: Backend> UNetBlock<B> for Upsample<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct DownsampleConfig {
     n_channels: usize,
 }
@@ -426,7 +426,7 @@ impl<B: Backend> UNetBlock<B> for Conv2d<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct SpatialTransformerConfig {
     n_channels: usize,
     n_context_state: usize,
@@ -480,7 +480,7 @@ impl<B: Backend> SpatialTransformer<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct TransformerBlockConfig {
     n_state: usize,
     n_context_state: usize,
@@ -526,7 +526,7 @@ impl<B: Backend> TransformerBlock<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct MLPConfig {
     n_state: usize,
     mult: usize,
@@ -554,7 +554,7 @@ impl<B: Backend> MLP<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct GEGLUConfig {
     n_state_in: usize,
     n_state_out: usize,
@@ -591,7 +591,7 @@ impl<B: Backend> GEGLU<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct MultiHeadAttentionConfig {
     n_state: usize,
     n_context_state: usize,
@@ -652,7 +652,7 @@ impl<B: Backend> MultiHeadAttention<B> {
     }
 }
 
-#[derive(Config)]
+#[derive(Config, Debug)]
 pub struct ResBlockConfig {
     n_channels_in: usize,
     n_channels_embed: usize,