From 7fadabea911684827008fe5d314cd121312ea1c1 Mon Sep 17 00:00:00 2001
From: "michal.kozlowski" <michalkozlowski936@gmail.com>
Date: Tue, 14 Mar 2023 21:13:02 +0100
Subject: [PATCH] train stream fix

---
 train_stream.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/train_stream.py b/train_stream.py
index 4e21818..6d83f2f 100644
--- a/train_stream.py
+++ b/train_stream.py
@@ -79,7 +79,7 @@ def main(config, hug_token):
             return obj
 
     def process(row, split):
-        task_start_token, prompt_end_token = "<s_cord-v2>"
+        task_start_token, prompt_end_token = "<s_cord-v2>", "<s_cord-v2>"
         ground_truth = json.loads(row["ground_truth"])
         if "gt_parses" in ground_truth:  # when multiple ground truths are available, e.g., docvqa
             assert isinstance(ground_truth["gt_parses"], list)