Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -4,16 +4,17 @@ from transformers import pipeline
|
|
| 4 |
|
| 5 |
pipe = pipeline("text-classification", model = "Pavan97/Emotion_Analysis" )
|
| 6 |
def predict(text):
|
| 7 |
-
|
| 8 |
tokenized_text = tokenizer(text, return_tensors = 'pt')
|
| 9 |
with torch.no_grad():
|
| 10 |
labels = model(**tokenized_text)
|
| 11 |
labels = torch.nn.functional.softmax(labels.logits[0], dim = -1)
|
| 12 |
|
| 13 |
-
predicted_labels =
|
| 14 |
for idx, label in enumerate(labels):
|
| 15 |
-
|
| 16 |
-
|
|
|
|
| 17 |
|
| 18 |
predicted_labels.sort(key = lambda label: label[1], reverse = True)
|
| 19 |
|
|
@@ -21,7 +22,7 @@ def predict(text):
|
|
| 21 |
|
| 22 |
iface = gr.Interface(fn = predict,
|
| 23 |
inputs = "text",
|
| 24 |
-
outputs =
|
| 25 |
title = 'Emotion Analysis',
|
| 26 |
description = "Get the emotion of the input text",)
|
| 27 |
|
|
|
|
| 4 |
|
| 5 |
pipe = pipeline("text-classification", model = "Pavan97/Emotion_Analysis" )
|
| 6 |
def predict(text):
|
| 7 |
+
|
| 8 |
tokenized_text = tokenizer(text, return_tensors = 'pt')
|
| 9 |
with torch.no_grad():
|
| 10 |
labels = model(**tokenized_text)
|
| 11 |
labels = torch.nn.functional.softmax(labels.logits[0], dim = -1)
|
| 12 |
|
| 13 |
+
predicted_labels = {}
|
| 14 |
for idx, label in enumerate(labels):
|
| 15 |
+
|
| 16 |
+
predicted_labels[id2label[str(idx)]] = label.numpy().item()
|
| 17 |
+
|
| 18 |
|
| 19 |
predicted_labels.sort(key = lambda label: label[1], reverse = True)
|
| 20 |
|
|
|
|
| 22 |
|
| 23 |
iface = gr.Interface(fn = predict,
|
| 24 |
inputs = "text",
|
| 25 |
+
outputs = gr.outputs.Label(num_top_classes=5),
|
| 26 |
title = 'Emotion Analysis',
|
| 27 |
description = "Get the emotion of the input text",)
|
| 28 |
|