.gitattributes

@@ -5,9 +5,6 @@
 <<<<<<< HEAD
 <<<<<<< HEAD
 <<<<<<< HEAD
-=======
-<<<<<<< HEAD
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.bin.* filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
 =======
@@ -21,15 +18,11 @@
 =======
 *.bz2 filter=lfs diff=lfs merge=lfs -text
 *.ckpt filter=lfs diff=lfs merge=lfs -text
-<<<<<<< HEAD
->>>>>>> 8283d5e4307c53a864411b01b43fc1d1365978ae
-=======
 >>>>>>> e4732774dd7bad94b1e2de2439cb8d7f55f65a7d
 =======
 *.bz2 filter=lfs diff=lfs merge=lfs -text
 *.ckpt filter=lfs diff=lfs merge=lfs -text
 >>>>>>> 57db79b3cf6873ca3c8d7d6b056500e4f5ac930f
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.ftz filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
@@ -39,22 +32,15 @@
 <<<<<<< HEAD
 <<<<<<< HEAD
 <<<<<<< HEAD
-=======
-<<<<<<< HEAD
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
 =======
 =======
 >>>>>>> ba5e8e17d2dab3c94db703030291c6349154e849
 =======
-<<<<<<< HEAD
->>>>>>> 8283d5e4307c53a864411b01b43fc1d1365978ae
-=======
 >>>>>>> e4732774dd7bad94b1e2de2439cb8d7f55f65a7d
 =======
 >>>>>>> 57db79b3cf6873ca3c8d7d6b056500e4f5ac930f
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
@@ -63,20 +49,13 @@
 <<<<<<< HEAD
 <<<<<<< HEAD
 <<<<<<< HEAD
-=======
-<<<<<<< HEAD
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 >>>>>>> 4e5a12c7333865beb434d4ee80b218c32d141352
 =======
 >>>>>>> ba5e8e17d2dab3c94db703030291c6349154e849
 =======
-<<<<<<< HEAD
->>>>>>> 8283d5e4307c53a864411b01b43fc1d1365978ae
-=======
 >>>>>>> e4732774dd7bad94b1e2de2439cb8d7f55f65a7d
 =======
 >>>>>>> 57db79b3cf6873ca3c8d7d6b056500e4f5ac930f
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text
@@ -85,9 +64,6 @@
 <<<<<<< HEAD
 <<<<<<< HEAD
 <<<<<<< HEAD
-=======
-<<<<<<< HEAD
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 *.rar filter=lfs diff=lfs merge=lfs -text
@@ -126,13 +102,9 @@ tokenizer.json filter=lfs diff=lfs merge=lfs -text
 =======
 >>>>>>> ba5e8e17d2dab3c94db703030291c6349154e849
 =======
-<<<<<<< HEAD
->>>>>>> 8283d5e4307c53a864411b01b43fc1d1365978ae
-=======
 >>>>>>> e4732774dd7bad94b1e2de2439cb8d7f55f65a7d
 =======
 >>>>>>> 57db79b3cf6873ca3c8d7d6b056500e4f5ac930f
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.pickle filter=lfs diff=lfs merge=lfs -text
 *.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
@@ -152,9 +124,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 <<<<<<< HEAD
 <<<<<<< HEAD
 <<<<<<< HEAD
-=======
-<<<<<<< HEAD
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40
 *.EncryptBy4pd filter=lfs diff=lfs merge=lfs -text
 >>>>>>> 4e5a12c7333865beb434d4ee80b218c32d141352
 =======
@@ -162,11 +131,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 >>>>>>> ba5e8e17d2dab3c94db703030291c6349154e849
 =======
 *.EncryptBy4pd filter=lfs diff=lfs merge=lfs -text
-<<<<<<< HEAD
->>>>>>> 8283d5e4307c53a864411b01b43fc1d1365978ae
-=======
 >>>>>>> e4732774dd7bad94b1e2de2439cb8d7f55f65a7d
 =======
 *.EncryptBy4pd filter=lfs diff=lfs merge=lfs -text
 >>>>>>> 57db79b3cf6873ca3c8d7d6b056500e4f5ac930f
->>>>>>> 29cbb4baae82dfd579ec44e76ad13d223dbd9c40

modeling_deepseekocr.py

@@ -945,6 +945,17 @@ class DeepseekOCRForCausalLM(DeepseekV2ForCausalLM):
                         use_cache = True
                         )
                 
+
+        if '<image>' in conversation[0]['content'] and eval_mode:
+                outputs = tokenizer.decode(output_ids[0, input_ids.unsqueeze(0).cuda().shape[1]:])
+                stop_str = '<｜end▁of▁sentence｜>'
+                if outputs.endswith(stop_str):
+                    outputs = outputs[:-len(stop_str)]
+                # re_match
+                outputs = outputs.strip()
+
+                return outputs
+        
         if '<image>' in conversation[0]['content'] and test_compress:
             outputs = tokenizer.decode(output_ids[0, input_ids.unsqueeze(0).cuda().shape[1]:])
             pure_texts_outputs_token_length = len(text_encode(tokenizer, outputs, bos=False, eos=False))
@@ -1024,13 +1035,3 @@ class DeepseekOCRForCausalLM(DeepseekV2ForCausalLM):
                 plt.close()
 
             result.save(f"{output_path}/result_with_boxes.jpg")
-
-        if '<image>' in conversation[0]['content'] and eval_mode:
-            outputs = tokenizer.decode(output_ids[0, input_ids.unsqueeze(0).cuda().shape[1]:])
-            stop_str = '<｜end▁of▁sentence｜>'
-            if outputs.endswith(stop_str):
-                outputs = outputs[:-len(stop_str)]
-            # re_match
-            outputs = outputs.strip()
-
-            return outputs