Add API server

app/app.py

@@ -0,0 +1,45 @@
+import argparse
+
+import flask
+from flask import Flask, jsonify, request
+import requests
+
+from utils import load_model, load_tokenizer, encode_text, predict
+from transformers import AutoTokenizer
+
+app = Flask(__name__)
+
+Model, config_file = load_model("./src/Baseline_model/state_dict.py", "./src/Baseline_model/config_model.json") # argparse
+tokenizer = load_tokenizer()
+
+@app.route('/predict', methods = ['POST'])
+def api_sentiment():
+    data = request.get_json(force = True)
+
+    encoded = encode_text(data['text'], tokenizer, config_file)
+
+    prediction = predict(Model, encoded)
+
+    request_responses = {
+        "Negative Score" : prediction[0][0],
+        "Positive Score" : prediction[0][1]
+    }
+
+    return jsonify(request_responses)
+
+if __name__ == "__main__":
+
+ """    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--path_to_model", help="Path to the pre trained model", typr=str)
+    parser.add_argument("--path_to_config", help="Path to the config file", type=str)
+
+    args = parser.parse_args()
+ """
+    app.run(debug=True)
+
+
+
+
+
+