Upload fixed & optimized fp16/q4f16 ONNX weights (#6)

- Upload fixed & optimized fp16/q4f16 ONNX weights (85cbfa5e8ce5091adafd6cc025a6e5346f8cedcf)
- Delete onnx/decoder_model_merged_fp16.onnx_data with huggingface_hub (4c5c62bdc94482e1baf3ee681060fc9bf793dd13)


Co-authored-by: Joshua <[email protected]>

onnx/decoder_model_merged_fp16.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3790c216622cad5b906bb01e8b82d21b88c1e0c756dd083ef6e736f13a784dc5
-size 2518137
+oid sha256:1baaffade2a1476da8aeebbfe03bd1b7f538efd09c9ad1967d8ae61f685c1eea
+size 1815458706

onnx/decoder_model_merged_fp16.onnx_data

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3a67fe6eec549caed51b6665b3f2d091cff8be922adc76748fc01b3a26cfa9f1
-size 3416289280

onnx/decoder_model_merged_q4f16.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb4ff19d4584948b1823a3a8aa57fe6973924c4a22bbe4fd0bd2729858343a43
-size 1157790536
+oid sha256:d36364b0bcc4f7db3ceea72e59018c28d67f105294a47436c954ea3f2b413055
+size 609597010