Keras
LiteRT
Devil-Assassin commited on
Commit
3a22027
·
verified ·
1 Parent(s): 6bfba57

Object Detection: Uploaded re-exported YOLO models (COCO)

Browse files

Re-exported the original YOLO models using the latest Ultralytics
export, which first modifies the model itself for full GPU delegation,
then uses litert-torch as the exporter.

I modified the export process so that I can export the models with NHWC
input instead of NCHW.

- COCO: Original YOLO models pretrained on COCO, exported to LiteRT.
- Precision: float32 and w8a32
- Resolution: 320x320 (letterboxed with value=(114, 114, 114))
- Input: 1x320x320x3 (NHWC)
- Output (det): 1x84x2100 (pre-NMS)
- Output (seg): 1x116x2100 (pre-NMS) and 1x32x80x80

object_detection/coco/yolo26n-seg_coco.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:361fbfabab285c3237700b6bb91d7ecfa602cd945fffda8dbe1242829b71e73f
3
+ size 6719965
object_detection/coco/yolo26n-seg_coco_f32.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66e8b86d213ea972b14e5af51751167ce135542b2d8e54030d72a304280da2ac
3
+ size 11241891
object_detection/coco/yolo26n-seg_coco_w8a32.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e386cd62f72cddcfe13ae1863e61f75c5ea9a71282045e8dd08c3dc6289dbaa8
3
+ size 3215630
object_detection/coco/yolo26n_coco.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b09cc8bf347f0fc8a5f7657480587f25db09b34bf33b0652110fb03a8ad4fef
3
+ size 5544453
object_detection/coco/yolo26n_coco_f32.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b794425ce6783b89996d3c2291028c4358a9fb86355c7ab43a074a09ba75b243
3
+ size 9944264
object_detection/coco/yolo26n_coco_w8a32.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f28837268a29cd6f9714b54624762bd694c30aa6196f14e8dc7b6652b550c73
3
+ size 2849400