Add mobileclip2

Author: jamjamjon

README.md

@@ -95,7 +95,7 @@ usls = "latest-version"
 | [CLIP](https://github.com/openai/CLIP) | Vision-Language Embedding | [demo](examples/clip) |
 | [jina-clip-v1](https://huggingface.co/jinaai/jina-clip-v1) | Vision-Language Embedding | [demo](examples/clip) |
 | [jina-clip-v2](https://huggingface.co/jinaai/jina-clip-v2) | Vision-Language Embedding | [demo](examples/clip) |
-| [mobileclip](https://github.com/apple/ml-mobileclip) | Vision-Language Embedding | [demo](examples/clip) |
+| [mobileclip & mobileclip2](https://github.com/apple/ml-mobileclip) | Vision-Language Embedding | [demo](examples/clip) |
 | [BLIP](https://github.com/salesforce/BLIP) | Image Captioning | [demo](examples/blip) |
 | [DB(PaddleOCR-Det)](https://arxiv.org/abs/1911.08947) | Text Detection | [demo](examples/db) |
 | [FAST](https://github.com/czczup/FAST) | Text Detection | [demo](examples/fast) |

examples/clip/README.md

@@ -3,7 +3,7 @@ This demo showcases how to use [CLIP](https://github.com/openai/CLIP) to compute
 ## Quick Start
 
 ```shell
-cargo run -r -F cuda --example clip -- --device cuda:0
+cargo run -r -F cuda --example clip -- --device cuda:0 --dtype q4f16
 ```
 
 ## Results

examples/clip/main.rs

@@ -22,8 +22,8 @@ fn main() -> Result<()> {
     let args: Args = argh::from_env();
 
     // build model
-    let config = Config::mobileclip_s0()
-        // mobileclip_blt()
+    let config = Config::mobileclip2_s0()
+        // mobileclip_s0()
         // clip_vit_b16()
         // clip_vit_l14()
         // clip_vit_b32()

src/models/clip/config.rs

@@ -108,4 +108,44 @@ impl crate::Config {
             .with_textual_file("blt-textual.onnx")
             .with_visual_file("blt-visual.onnx")
     }
+
+    pub fn mobileclip2() -> Self {
+        Self::mobileclip().with_name("mobileclip2")
+    }
+
+    pub fn mobileclip2_s0() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("s0-textual.onnx")
+            .with_visual_file("s0-visual.onnx")
+    }
+
+    pub fn mobileclip2_s2() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("s2-textual.onnx")
+            .with_visual_file("s2-visual.onnx")
+    }
+
+    pub fn mobileclip2_s3() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("s3-textual.onnx")
+            .with_visual_file("s3-visual.onnx")
+    }
+
+    pub fn mobileclip2_s4() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("s4-textual.onnx")
+            .with_visual_file("s4-visual.onnx")
+    }
+
+    pub fn mobileclip2_b() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("b-textual.onnx")
+            .with_visual_file("b-visual.onnx")
+    }
+
+    pub fn mobileclip2_l_14() -> Self {
+        Self::mobileclip2()
+            .with_textual_file("l-14-textual.onnx")
+            .with_visual_file("l-14-visual.onnx")
+    }
 }