# DiT outputs clarification

**URL:** <https://discuss.huggingface.co/t/dit-outputs-clarification/49149>\
**Category:** Models\
**Created:** [August 2, 2023, 1:37pm UTC](https://discuss.huggingface.co/t/dit-outputs-clarification/49149 "2023-08-02T13:37:31Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Francesco](https://avatars.discourse-cdn.com/v4/letter/f/e47c2d/32.png) [@Francesco](https://discuss.huggingface.co/u/Francesco)\
**Post date:** [August 2, 2023, 1:37pm UTC](https://discuss.huggingface.co/t/dit-outputs-clarification/49149/1 "2023-08-02T13:37:31Z")

</div>

Hi there!

Trying to run DiT but I cannot see any doc about what are the outputs and what I should do with them, example

```python
from transformers import BeitImageProcessor, BeitForMaskedImageModeling
import torch
from PIL import Image

image = Image.open('path_to_your_document_image').convert('RGB')

processor = BeitImageProcessor.from_pretrained("microsoft/dit-large")
model = BeitForMaskedImageModeling.from_pretrained("microsoft/dit-large")

num_patches = (model.config.image_size // model.config.patch_size) ** 2
pixel_values = processor(images=image, return_tensors="pt").pixel_values
# create random boolean mask of shape (batch_size, num_patches)
bool_masked_pos = torch.randint(low=0, high=2, size=(1, num_patches)).bool()

outputs = model(pixel_values, bool_masked_pos=bool_masked_pos)
loss, logits = outputs.loss, outputs.logits

```

could you provide more clarification about logit? Happy to PR a fix for the doc later

Thanks