t

app.py

@@ -18,41 +18,20 @@ import gradio as gr
 # except Exception as e:
 #     ex = e
 
+import requests
 def f(text):
-    import requests
-
-    API_URL = "https://api-inference.huggingface.co/models/EleutherAI/gpt-j-6B"
-    headers = {"Authorization": "Bearer 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"}
-
-    def query(payload):
-        response = requests.post(API_URL, headers=headers, json=payload)
-        return response.json()
-
-    output = query(text)
-    return output
-
-    # try:
-    #     if ex:
-    #         raise Exception("err from transformers import: \n"+str(ex))
-    #     model = AutoModelForCausalLM.from_pretrained("EleutherAI/gpt-j-6B")
-    #     default_do_sample = False #True (default is True)
-    #     input_ids = tokenizer(text, return_tensors="pt").input_ids
-    #     gen_tokens = model.generate(input_ids, do_sample=default_do_sample, temperature=0.9, max_length=30)
-    #     gen_text = tokenizer.batch_decode(gen_tokens)[0]
-    #     return gen_text
-    # except Exception as e:
-    #     return "err: \n" + str(e)
-    # from transformers import AutoModelForCausalLM, AutoTokenizer
-    # return text
-# def predict(text):
-    # from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    # tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-j-6B")
-#     model = AutoModelForCausalLM.from_pretrained("EleutherAI/gpt-j-6B")
-#     input_ids = tokenizer(text, return_tensors="pt").input_ids
-#     gen_tokens = model.generate(input_ids, do_sample=True, temperature=0.9, max_length=100)
-#     gen_text = tokenizer.batch_decode(gen_tokens)[0]
-#     return gen_text
+    context = text
+    payload = {
+        "context": context,
+        "token_max_length": 512,
+        "temperature": 1.0,
+        "top_p": 0.9,
+    }
+    response = requests.post("http://api.vicgalle.net:5000/generate", params=payload).json()
+    # http://api.vicgalle.net:5000/docs#/default/generate_generate_post
+    return response.text
+
+   
 
 iface = gr.Interface(fn=f, inputs="text", outputs="text")
 iface.launch(enable_queue=True)