Update app.py
Browse files
app.py
CHANGED
@@ -1,24 +1,24 @@
|
|
1 |
from transformers import AutoFeatureExtractor, ResNetForImageClassification
|
2 |
import torch
|
3 |
-
from datasets import load_dataset
|
4 |
|
5 |
-
dataset = load_dataset("huggingface/cats-image")
|
6 |
-
image = dataset["test"]["image"][0]
|
7 |
|
8 |
feature_extractor = AutoFeatureExtractor.from_pretrained("microsoft/resnet-50")
|
9 |
model = ResNetForImageClassification.from_pretrained("microsoft/resnet-50")
|
10 |
|
11 |
-
inputs = feature_extractor(image, return_tensors="pt")
|
12 |
-
|
13 |
-
with torch.no_grad():
|
14 |
-
logits = model(**inputs).logits
|
15 |
-
|
16 |
-
# model predicts one of the 1000 ImageNet classes
|
17 |
-
predicted_label = logits.argmax(-1).item()
|
18 |
-
print(model.config.id2label[predicted_label])
|
19 |
-
|
20 |
import gradio as gr
|
21 |
def segment(image):
|
22 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
|
24 |
-
gr.Interface(fn=segment, inputs="image", outputs="
|
|
|
1 |
from transformers import AutoFeatureExtractor, ResNetForImageClassification
|
2 |
import torch
|
3 |
+
# from datasets import load_dataset
|
4 |
|
5 |
+
# dataset = load_dataset("huggingface/cats-image")
|
6 |
+
# image = dataset["test"]["image"][0]
|
7 |
|
8 |
feature_extractor = AutoFeatureExtractor.from_pretrained("microsoft/resnet-50")
|
9 |
model = ResNetForImageClassification.from_pretrained("microsoft/resnet-50")
|
10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
import gradio as gr
|
12 |
def segment(image):
|
13 |
+
inputs = feature_extractor(image, return_tensors="pt")
|
14 |
+
|
15 |
+
with torch.no_grad():
|
16 |
+
logits = model(**inputs).logits
|
17 |
+
|
18 |
+
# model predicts one of the 1000 ImageNet classes
|
19 |
+
predicted_label = logits.argmax(-1).item()
|
20 |
+
# print(model.config.id2label[predicted_label])
|
21 |
+
|
22 |
+
return model.config.id2label[predicted_label]
|
23 |
|
24 |
+
gr.Interface(fn=segment, inputs="image", outputs="label").launch()
|