update format

src/models/mod.rs

@@ -1,7 +1,9 @@
 pub mod qwen2_5vl;
 use anyhow::Result;
 use candle_core::{DType, Device};
-use openai_dive::v1::resources::chat::{ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse};
+use openai_dive::v1::resources::chat::{
+    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
+};
 use rocket::futures::Stream;
 
 pub trait GenerateModel {
@@ -9,6 +11,8 @@ pub trait GenerateModel {
     where
         Self: Sized;
     fn generate(&mut self, mes: ChatCompletionParameters) -> Result<ChatCompletionResponse>;
-    fn generate_stream(&mut self, mes: ChatCompletionParameters) -> Result<impl Stream<Item = Result<ChatCompletionChunkResponse, anyhow::Error>>>;
+    fn generate_stream(
+        &mut self,
+        mes: ChatCompletionParameters,
+    ) -> Result<impl Stream<Item = Result<ChatCompletionChunkResponse, anyhow::Error>>>;
 }
-

src/models/qwen2_5vl/mod.rs

@@ -1,4 +1,4 @@
 pub mod config;
+pub mod generate;
 pub mod model;
 pub mod processor;
-pub mod generate;

src/tokenizer/mod.rs

@@ -1 +1 @@
-pub mod tokenizer;
+pub mod tokenizer;

src/utils/audio_utils.rs

@@ -0,0 +1 @@
+

src/utils/mod.rs

@@ -1,4 +1,4 @@
+pub mod img_utils;
 pub mod tensor_utils;
 pub mod utils;
-pub mod img_utils;
-pub mod video_utils;
+pub mod video_utils;

src/utils/utils.rs

@@ -3,7 +3,9 @@ use candle_core::{DType, Device};
 use candle_transformers::generation::LogitsProcessor;
 use openai_dive::v1::resources::{
     chat::{
-        ChatCompletionChoice, ChatCompletionChunkChoice, ChatCompletionChunkResponse, ChatCompletionResponse, ChatMessage, ChatMessageContent, DeltaChatMessage, DeltaFunction, DeltaToolCall, Function, ToolCall
+        ChatCompletionChoice, ChatCompletionChunkChoice, ChatCompletionChunkResponse,
+        ChatCompletionResponse, ChatMessage, ChatMessageContent, DeltaChatMessage, DeltaFunction,
+        DeltaToolCall, Function, ToolCall,
     },
     shared::FinishReason,
 };