v0.40.0

See https://github.com/quic/ai-hub-models/releases/v0.40.0 for changelog.

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+DEPLOYMENT_MODEL_LICENSE.pdf filter=lfs diff=lfs merge=lfs -text

DeformableDETR.onnx → DEPLOYMENT_MODEL_LICENSE.pdf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d2f357cb5f9a70843ad63216dcee89f1b6f635ac9ab4892e00257d2e19f9ccd
-size 165001956
+oid sha256:4409f93b0e82531303b3e10f52f1fdfb56467a25f05b7441c6bbd8bb8a64b42c
+size 109629

DeformableDETR_float.onnx.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a87dcb2c402f7e200dd3929d352272ac36c9be42123f96724b65ce3146494f43
+size 152084105

LICENSE

@@ -0,0 +1,2 @@
+The license of the original trained model can be found at https://github.com/facebookresearch/detr/blob/main/LICENSE.
+The license for the deployable model files (.tflite, .onnx, .dlc, .bin, etc.) can be found in DEPLOYMENT_MODEL_LICENSE.pdf.

README.md

@@ -1,6 +1,6 @@
 ---
 library_name: pytorch
-license: apache-2.0
+license: other
 tags:
 - android
 pipeline_tag: object-detection
@@ -23,21 +23,22 @@ More details on model performance across various devices, can be found
 [here](https://aihub.qualcomm.com/models/deformable_detr).
 
 
+
 ### Model Details
 
-- **Model Type:** Object detection
+- **Model Type:** Model_use_case.object_detection
 - **Model Stats:**
   - Model checkpoint: deformable-detr
   - Input resolution: 480x480
   - Number of parameters: 40M
   - Model size: 160 MB
 
-| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
+| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
-| DeformableDETR | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 995.627 ms | 31 - 98 MB | FP16 | NPU | [DeformableDETR.onnx](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx) |
-| DeformableDETR | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 785.492 ms | 44 - 150 MB | FP16 | NPU | [DeformableDETR.onnx](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx) |
-| DeformableDETR | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 644.37 ms | 45 - 161 MB | FP16 | NPU | [DeformableDETR.onnx](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx) |
-| DeformableDETR | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1193.916 ms | 106 - 106 MB | FP16 | NPU | [DeformableDETR.onnx](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx) |
+| DeformableDETR | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1255.749 ms | 85 - 314 MB | NPU | [DeformableDETR.onnx.zip](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx.zip) |
+| DeformableDETR | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 922.184 ms | 158 - 412 MB | NPU | [DeformableDETR.onnx.zip](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx.zip) |
+| DeformableDETR | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 820.606 ms | 169 - 421 MB | NPU | [DeformableDETR.onnx.zip](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx.zip) |
+| DeformableDETR | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1483.574 ms | 132 - 132 MB | NPU | [DeformableDETR.onnx.zip](https://huggingface.co/qualcomm/DeformableDETR/blob/main/DeformableDETR.onnx.zip) |
 
 
 
@@ -95,17 +96,7 @@ device. This script does the following:
 ```bash
 python -m qai_hub_models.models.deformable_detr.export
 ```
-```
-Profiling Results
-------------------------------------------------------------
-DeformableDETR
-Device                          : Samsung Galaxy S23 (13)     
-Runtime                         : ONNX                        
-Estimated inference time (ms)   : 995.6                       
-Estimated peak memory usage (MB): [31, 98]                    
-Total # Ops                     : 1360                        
-Compute Unit(s)                 : NPU (1238 ops) CPU (122 ops)
-```
+
 
 
 ## How does this work?
@@ -129,7 +120,7 @@ from qai_hub_models.models.deformable_detr import Model
 torch_model = Model.from_pretrained()
 
 # Device
-device = hub.Device("Samsung Galaxy S24")
+device = hub.Device("Samsung Galaxy S25")
 
 # Trace model
 input_shape = torch_model.get_input_spec()
@@ -191,13 +182,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
 You can also run the demo on-device.
 
 ```bash
-python -m qai_hub_models.models.deformable_detr.demo --on-device
+python -m qai_hub_models.models.deformable_detr.demo --eval-mode on-device
 ```
 
 **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
 environment, please add the following to your cell (instead of the above).
 ```
-%run -m qai_hub_models.models.deformable_detr.demo -- --on-device
+%run -m qai_hub_models.models.deformable_detr.demo -- --eval-mode on-device
 ```
 
 

tool-versions.yaml

@@ -0,0 +1,4 @@
+tool_versions:
+  onnx:
+    qairt: 2.37.1.250807093845_124904
+    onnx_runtime: 1.23.0