📰 Update about.md

packages/tasks/src/tasks/depth-estimation/about.md

@@ -24,7 +24,7 @@ With the `transformers` library, you can use the `depth-estimation` pipeline to
 ```python
 from transformers import pipeline
 
-estimator = pipeline(task="depth-estimation", model="Intel/dpt-large")
+estimator = pipeline(task="depth-estimation", model="Intel/dpt-large", device=0)
 result = estimator(images="http://images.cocodataset.org/val2017/000000039769.jpg")
 result
 

packages/tasks/src/tasks/image-classification/about.md

@@ -16,12 +16,12 @@ With the `transformers` library, you can use the `image-classification` pipeline
 
 ```python
 from transformers import pipeline
-clf = pipeline("image-classification")
-clf("path_to_a_cat_image")
+image_classifier = pipeline(task="image-classification", model="microsoft/resnet-50", device=0)
+result = image_classifier(IMAGE_PATH)
+result
 
-[{'label': 'tabby cat', 'score': 0.731},
-...
-]
+# [{'label': 'Egyptian cat', 'score': 0.7005051970481873},
+# {'label': 'tabby, tabby cat', 'score': 0.16163259744644165}]
 ```
 
 You can use [huggingface.js](https://github.com/huggingface/huggingface.js) to classify images using models on Hugging Face Hub.

packages/tasks/src/tasks/image-segmentation/about.md

@@ -30,10 +30,17 @@ Panoptic Segmentation is the Image Segmentation task that segments the image bot
 
 You can infer with Image Segmentation models using the `image-segmentation` pipeline. You need to install [timm](https://github.com/rwightman/pytorch-image-models) first.
 
-```python
+```bash
 !pip install timm
-model = pipeline("image-segmentation")
-model("cat.png")
+```
+
+```python
+from transformers import pipeline
+
+model = pipeline(task="image-segmentation", model="facebook/detr-resnet-50-panoptic", device=0)
+result = model(IMAGE_PATH)
+result
+
 #[{'label': 'cat',
 #  'mask': mask_code,
 #  'score': 0.999}

packages/tasks/src/tasks/object-detection/about.md

@@ -21,9 +21,9 @@ Object Detection models are used to count instances of objects in a given image,
 You can infer with Object Detection models through the `object-detection` pipeline. When calling the pipeline you just need to specify a path or http link to an image.
 
 ```python
-model = pipeline("object-detection")
-
-model("path_to_cat_image")
+model = pipeline(task="object-detection", model="facebook/detr-resnet-50", device=0)
+result = model(IMAGE_PATH)
+result
 
 # [{'label': 'blanket',
 #  'mask': mask_string,

packages/tasks/src/tasks/video-classification/about.md

@@ -17,7 +17,7 @@ Below you can find code for inferring with a pre-trained video classification mo
 ```python
 from transformers import pipeline
 
-pipe = pipeline(task = "video-classification", model="nateraw/videomae-base-finetuned-ucf101-subset")
+pipe = pipeline(task="video-classification", model="nateraw/videomae-base-finetuned-ucf101-subset", device=0)
 pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/basketball.avi?download=true")
 
 #[{'score': 0.90, 'label': 'BasketballDunk'},

packages/tasks/src/tasks/zero-shot-classification/about.md

@@ -26,7 +26,7 @@ You can use the 🤗 Transformers library zero-shot-classification pipeline to i
 ```python
 from transformers import pipeline
 
-pipe = pipeline(model="facebook/bart-large-mnli")
+pipe = pipeline(task="zero-shot-classification", model="facebook/bart-large-mnli")
 pipe("I have a problem with my iphone that needs to be resolved asap!",
     candidate_labels=["urgent", "not urgent", "phone", "tablet", "computer"],
 )

packages/tasks/src/tasks/zero-shot-image-classification/about.md

@@ -32,7 +32,8 @@ The model can be loaded with the zero-shot-image-classification pipeline like so
 from transformers import pipeline
 # More models in the model hub.
 model_name = "openai/clip-vit-large-patch14-336"
-classifier = pipeline("zero-shot-image-classification", model = model_name)
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+classifier = pipeline(task="zero-shot-image-classification", model=model_name, device=device)
 ```
 
 You can then use this pipeline to classify images into any of the class names you specify. You can specify more than two class labels too.

packages/tasks/src/tasks/zero-shot-object-detection/about.md

@@ -24,7 +24,7 @@ from PIL import Image
 
 image = Image.open("my-image.png").convert("RGB")
 
-detector = pipeline(model="google/owlvit-base-patch32", task="zero-shot-object-detection")
+detector = pipeline(task="zero-shot-object-detection", model="google/owlvit-base-patch32")
 
 predictions = detector(
     image,