nits
Browse files
README.md
CHANGED
|
@@ -40,17 +40,18 @@ Below we used [`"llava-hf/llava-interleave-qwen-0.5b-hf"`](https://huggingface.c
|
|
| 40 |
|
| 41 |
|
| 42 |
```python
|
| 43 |
-
from transformers import pipeline
|
| 44 |
from PIL import Image
|
| 45 |
import requests
|
| 46 |
|
| 47 |
model_id = "llava-hf/llava-interleave-qwen-7b-hf"
|
| 48 |
pipe = pipeline("image-to-text", model=model_id)
|
|
|
|
| 49 |
|
| 50 |
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
|
| 51 |
image = Image.open(requests.get(url, stream=True).raw)
|
| 52 |
|
| 53 |
-
# Define a chat
|
| 54 |
# Each value in "content" has to be a list of dicts with types ("text", "image")
|
| 55 |
conversation = [
|
| 56 |
{
|
|
@@ -88,7 +89,7 @@ model = LlavaForConditionalGeneration.from_pretrained(
|
|
| 88 |
|
| 89 |
processor = AutoProcessor.from_pretrained(model_id)
|
| 90 |
|
| 91 |
-
# Define a chat
|
| 92 |
# Each value in "content" has to be a list of dicts with types ("text", "image")
|
| 93 |
conversation = [
|
| 94 |
{
|
|
|
|
| 40 |
|
| 41 |
|
| 42 |
```python
|
| 43 |
+
from transformers import pipeline, AutoProcessor
|
| 44 |
from PIL import Image
|
| 45 |
import requests
|
| 46 |
|
| 47 |
model_id = "llava-hf/llava-interleave-qwen-7b-hf"
|
| 48 |
pipe = pipeline("image-to-text", model=model_id)
|
| 49 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 50 |
|
| 51 |
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
|
| 52 |
image = Image.open(requests.get(url, stream=True).raw)
|
| 53 |
|
| 54 |
+
# Define a chat history and use `apply_chat_template` to get correctly formatted prompt
|
| 55 |
# Each value in "content" has to be a list of dicts with types ("text", "image")
|
| 56 |
conversation = [
|
| 57 |
{
|
|
|
|
| 89 |
|
| 90 |
processor = AutoProcessor.from_pretrained(model_id)
|
| 91 |
|
| 92 |
+
# Define a chat history and use `apply_chat_template` to get correctly formatted prompt
|
| 93 |
# Each value in "content" has to be a list of dicts with types ("text", "image")
|
| 94 |
conversation = [
|
| 95 |
{
|