Spaces:
Running
Running
Delete app.py
Browse files
app.py
DELETED
@@ -1,34 +0,0 @@
|
|
1 |
-
from transformers import AutoModel, AutoTokenizer
|
2 |
-
from flask import Flask, request, jsonify
|
3 |
-
import tensorflow as tf
|
4 |
-
|
5 |
-
app = Flask(__name__)
|
6 |
-
|
7 |
-
# Load Hugging Face model and tokenizer
|
8 |
-
tokenizer = AutoTokenizer.from_pretrained("Erfan11/Neuracraft", use_auth_token="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")
|
9 |
-
hf_model = AutoModel.from_pretrained("Erfan11/Neuracraft", use_auth_token="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")
|
10 |
-
|
11 |
-
# Load TensorFlow model
|
12 |
-
tf_model = tf.keras.models.load_model('path_to_your_tf_model.h5')
|
13 |
-
|
14 |
-
@app.route('/predict', methods=['POST'])
|
15 |
-
def predict():
|
16 |
-
data = request.get_json()
|
17 |
-
|
18 |
-
# Tokenize the input using Hugging Face's tokenizer
|
19 |
-
inputs = tokenizer(data["text"], return_tensors="pt")
|
20 |
-
|
21 |
-
# Make prediction with Hugging Face model
|
22 |
-
hf_outputs = hf_model(**inputs)
|
23 |
-
|
24 |
-
# Optionally: You can also add TensorFlow model predictions here, depending on what it’s used for.
|
25 |
-
# Assuming the TensorFlow model is used for something else like feature extraction
|
26 |
-
tf_outputs = tf_model.predict([data["text"]]) # Modify based on your input processing
|
27 |
-
|
28 |
-
return jsonify({
|
29 |
-
"hf_outputs": hf_outputs[0].tolist(), # Convert Hugging Face output to JSON serializable format
|
30 |
-
"tf_outputs": tf_outputs.tolist() # Convert TensorFlow output to JSON serializable format
|
31 |
-
})
|
32 |
-
|
33 |
-
if __name__ == '__main__':
|
34 |
-
app.run(debug=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|