Fix visual bug in quantization (#92)

It didn't calculate the compression ratio properly because of a
copy/paste error :(
This commit is contained in:
LoganDark 2023-06-07 04:45:21 -07:00 committed by GitHub
parent 5b41cd7e5d
commit 09ec3145b3
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions

View File

@ -1264,7 +1264,7 @@ bool rwkv_quantize_model_file(const char * in_path, const char * out_path, const
RWKV_ASSERT_FALSE_MSG(RWKV_ERROR_FILE_WRITE, rwkv_fwrite_tensor(out_file, tensor), "Failed to write tensor %s", name_str);
orig_total_size += orig_size;
new_total_size += orig_size;
new_total_size += new_size;
}
RWKV_MSG("original size = %8.2f MB\n", orig_total_size / 1024.0 / 1024.0);