Update README.md

README.md

@@ -25,12 +25,41 @@ Please place the downloaded PDF files in the ./pdf directory.
 - Copy jgraphqa.yaml, utils.py, and the generated jgraphqa.parquet file into the [lmms_eval/tasks](https://github.com/EvolvingLMMs-Lab/lmms-eval/tree/main/lmms_eval/tasks)/jgraphqa directory.
 (You will need to create the jgraphqa directory if it does not already exist.)
 - Please modify [lmms_eval/models/llava_onevision.py](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py) as follows.
- - Please add the following code below line [284](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py#L284).
+  - Please add the following code below line [284](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py#L284).
 ```python
+                    # image_tensor = process_images(visual, self._image_processor, self._config)　# This part is already present in the original code
+                    inputs = self._image_processor(visual)
+                    image_tensor = torch.tensor(inputs['pixel_values']).to(dtype=torch.float16, device=self.device)
+                    image_tensor = [image_tensor]
+                    # if type(image_tensor) is list:
+                    #     image_tensor = [_image.to(dtype=torch.float16, device=self.device) for _image in image_tensor]
+                    # else:
+                    #     image_tensor = image_tensor.to(dtype=torch.float16, device=self.device)
 
-
-
+                    task_type = "image"　# This part is already present in the original code
 ```
+  - Please add the following code below line [342](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py#L342).
+```python
+                kwargs["image_sizes"] = [[v.size[0], v.size[1]] for v in visual] if isinstance(visual, list) else [[visual.size[0], visual.size[1]]]　# This part is already present in the original code
+                _image_grid_thw = torch.tensor(inputs['image_grid_thw'], dtype=torch.long)
+                kwargs["image_grid_thws"] = [_image_grid_thw]
+            elif task_type == "video":　# This part is already present in the original code
+```
+  - Please add the following code below line [455](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py#L455).
+```python
+                        # image_tensor = process_images(visual, self._image_processor, self._config)　# This part is already present in the original code
+                        inputs = self._image_processor(visual)
+                        image_tensor = torch.tensor(inputs['pixel_values']).to(dtype=torch.float16, device=self.device)
+                        image_tensor = [image_tensor]
+                        if type(image_tensor) is list:　# This part is already present in the original code
+```
+  - Please add the following code below line [539](https://github.com/EvolvingLMMs-Lab/lmms-eval/blob/main/lmms_eval/models/llava_onevision.py#L539).
+```python
+                gen_kwargs["image_sizes"] = [batched_visuals[0][idx].size for idx in range(len(batched_visuals[0]))]　# This part is already present in the original code
+                _image_grid_thw = torch.tensor(inputs['image_grid_thw'], dtype=torch.long)
+                gen_kwargs["image_grid_thws"] = [_image_grid_thw]
+```
+
 
 ## 🤗Usage
 - Using the lmms-eval framework, please run the following command: