From: kunnis <redacted>
Date: Mon, 8 Apr 2024 15:44:19 +0000 (-0500)
Subject: Comment explaining a decision (#6531)
X-Git-Tag: upstream/0.0.4488~1854
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=cecd8d3c98b48f51aaa1d4c729e55bd319f6799c;p=pkg%2Fggml%2Fsources%2Fllama.cpp

Comment explaining a decision (#6531)
---

diff --git a/convert.py b/convert.py
index 244eb758..a37aeb5e 100755
--- a/convert.py
+++ b/convert.py
@@ -139,7 +139,8 @@ class GGMLFileType(enum.IntEnum):
         dt = GGML_FILE_TYPE_TO_DATA_TYPE.get(self)
         if dt is None:
             raise ValueError(self)
-        # 1D tensors are always F32.
+        # Convert all 1D tensors to F32.  Most of the codebase that takes in 1D tensors only handles F32 tensors, and most of the outputs tensors are F32.
+        #  Also The 1d tensors aren't much of a performance/size issue.  So instead of having to have separate F32 and F16 implementations of both, just convert everything to F32 for now.
         return dt if len(tensor.shape) > 1 else DT_F32