nlp_hero added

nlp_hero.py

@@ -0,0 +1,196 @@
+import openai
+import gradio as gr
+from pytube import YouTube
+from youtube_dl import YoutubeDL
+import yt_dlp
+import os
+from pydub import AudioSegment
+import time
+
+#openai.api_key = os.environ.get("OPENAI_API_KEY")
+
+conversation_history=[]
+
+def initial_prompt():
+
+    #openai.api_key = os.environ.get("OPENAI_API_KEY")
+
+    global conversation_history
+    conversation_history.append(f"""Türkçe doğal dil işleme fonksiyonlarını yerine getiren başarılı ve yetenekli bir dil modeli gibi davran.
+                                Yeteneklerin şu şekilde : [
+                                "1: Senden duygu analizi yapman istendiğinde cümlelerin pozitif mi yoksa negatif mi olduğunu yüzdesel değeri ile söyleyeceksin ve 
+                                sebebini açıklayacaksın.",
+                                "2: Cümlenin NER'lerini çıkarman yada ayırman istendiğinde o cümleleri NER'lerine ayırarak ve NER tipini belirterek geri döneceksin. Argo ve küfür metinlerini de NER listene dahil edeceksin.",
+                                "3: Sana verilen metnin özeti istendiğinde özetini çıkaracaksın.",
+                                "4: Cümlenin kategorik sınıflandırılması istendiğinde o cümlenin yada metnin ["magazin", "spor", "politika", "ekonomi", "eğitim", "turizm"] sektörlerinden 
+                                en yüksek olasılığa sahip olanı yüzdesel değeri ile beraber döneceksin.",
+                                "5: Verilen metnin anafikri istendiğinde, o metinde asıl anlatılmak istenenin ne olduğunu açıklayarak geri döneceksin."
+                                ]
+                                Argo ve küfürlü cümlelerinde analizini yaparak geri döneceksin.
+                                """)
+    
+    #"6: Metnin bütün dil özelliklerini çıkarmanı istediğimde önceki 5 maddeyi sırasıyla uygulayacak ve sonuclarını döneceksin."
+    
+    while True:
+
+        try:
+            completion = openai.ChatCompletion.create(
+                model="gpt-4", 
+                messages=[{"role": "system", "content": " ".join([str(item) for item in conversation_history])}],
+                temperature=0.7,
+                max_tokens=7500,
+            )
+
+            message = completion.choices[0].message['content']
+            print(message)
+
+            break
+
+        except Exception as e:
+            time.sleep(10)
+            continue
+
+    #message = completion.choices[0].message['content']
+    print(message)
+
+
+#initial_prompt()
+
+def handle_input(
+                input_str : str,
+                temperature,
+                max_tokens,
+                top_p_input,
+                presence_penalty,
+                frequency_penalty
+                ):
+    global conversation_history
+
+    if len(conversation_history) == 0:
+        initial_prompt()
+
+    if len(conversation_history) >=3:
+        conversation_history = conversation_history[:1] #+ conversation_history[-5:]
+
+    conversation_history.append(f"{input_str}\n")
+    
+    content = " ".join([str(item) for item in conversation_history])
+
+    while True:
+
+        try:
+
+            completion = openai.ChatCompletion.create(
+            model="gpt-4", 
+            messages=[{"role": "assistant", "content": content}],
+            temperature=temperature,
+            max_tokens=max_tokens,
+            presence_penalty=presence_penalty,
+            frequency_penalty=frequency_penalty,
+            top_p = top_p_input,
+            #stream = stream_input
+            )
+
+            message = completion.choices[0].message['content']
+            break
+
+        except Exception as e: 
+            time.sleep(5)
+            continue
+
+    conversation_history.append(f"{message}\n")
+    
+    return message
+
+def get_model_reply(query,api_key,temperature, max_tokens,top_p_input, presence_penalty, frequency_penalty,context=[]):
+ 
+    context += [query]
+    
+    openai.api_key = api_key
+
+    response = handle_input(query, temperature, max_tokens,top_p_input, presence_penalty, frequency_penalty)
+
+    
+    context += [response]
+    
+    responses = [(u,b) for u,b in zip(context[::2], context[1::2])]
+    
+    return responses, context
+
+
+def speech_2_text(audio,api_key, context=[]):
+    openai.api_key = api_key
+    audio_file= open(audio, "rb")
+    transcript = openai.Audio.transcribe("whisper-1", audio_file)
+    prompt = transcript.text
+    context += [prompt]
+
+    completion = openai.ChatCompletion.create(
+        model="gpt-4", 
+        messages=[{"role": "user", "content": prompt}],
+        max_tokens=7500,
+    )
+
+    response = completion.choices[0].message['content']
+    context += [response]
+    
+    responses = [(u,b) for u,b in zip(context[::2], context[1::2])]
+
+    return responses, context
+
+def get_audio(url,api_key,context=[]):
+    openai.api_key = api_key
+    destination = f"./audio/test"
+    try:
+        ydl_opts = {
+            'format': 'bestaudio/best',
+            'outtmpl':destination,
+            'postprocessors': [{
+                'key': 'FFmpegExtractAudio',
+                'preferredcodec': 'mp3',
+                'preferredquality': '192',
+            }],
+        }
+        with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+            ydl.download([url])
+        
+        audio_file= open(f'{destination}.mp3', "rb")
+        transcript = openai.Audio.transcribe("whisper-1", audio_file)
+
+        context += [transcript.text]
+        
+        responses = [(u,b) for u,b in zip(context[::2], context[1::2])]
+
+        return responses, context
+    
+    except Exception as e:
+        print("Connection Error")
+
+
+with gr.Blocks(css="#chatbot {overflow-y:auto; min-height:400px;} ") as dialog_app:
+
+    with gr.Tab("ChatBot"):
+        with gr.Row():
+            with gr.Column(scale=1, min_width=600):
+                api_key = gr.Textbox(label="Your API Key", type="password")
+                temperature_input = gr.inputs.Slider(minimum=0, maximum=1.0, default=0.5, step=0.01, label="Temperature")
+                max_tokens_input = gr.inputs.Slider(minimum=1, maximum=7500, default=50, step=10, label="Max Tokens")
+                top_p_input = gr.inputs.Slider(minimum=0.1, maximum=1.0, default=0.5, step=0.01, label="Top P")
+                presence_penalty_input = gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.0, step=0.1, label="Presence Penalty")
+                frequency_penalty_input = gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.0, step=0.1, label="Frequency Penalty")
+                #stream_input = gr.inputs.Checkbox(label="Stream")
+
+            with gr.Column(scale=2, min_width=600):
+                chatbot = gr.Chatbot(elem_id="chatbot")
+                state = gr.State([])
+
+                txt = gr.Textbox(
+                    show_label=False, 
+                    placeholder="Enter text and press enter"
+                ).style(container=False)
+
+        txt.submit(get_model_reply, [txt, api_key, temperature_input, max_tokens_input,top_p_input, presence_penalty_input,frequency_penalty_input, state], [chatbot, state])
+        txt.submit(lambda :"", None, txt)
+    
+
+dialog_app.launch()

requirements.txt

@@ -0,0 +1,76 @@
+aiofiles==23.1.0
+aiohttp==3.8.4
+aiosignal==1.3.1
+altair==4.2.2
+anyio==3.6.2
+async-timeout==4.0.2
+attrs==22.2.0
+Brotli==1.0.9
+certifi @ file:///C:/b/abs_85o_6fm0se/croot/certifi_1671487778835/work/certifi
+charset-normalizer==3.1.0
+click==8.1.3
+colorama==0.4.6
+contourpy==1.0.7
+cycler==0.11.0
+entrypoints==0.4
+fastapi==0.95.0
+ffmpy==0.3.0
+filelock==3.10.7
+fonttools==4.39.2
+frozenlist==1.3.3
+fsspec==2023.3.0
+gradio==3.23.0
+h11==0.14.0
+httpcore==0.16.3
+httpx==0.23.3
+huggingface-hub==0.13.3
+idna==3.4
+importlib-resources==5.12.0
+Jinja2==3.1.2
+jsonschema==4.17.3
+keras-tuner==1.2.1
+kiwisolver==1.4.4
+linkify-it-py==2.0.0
+markdown-it-py==2.2.0
+MarkupSafe==2.1.2
+matplotlib==3.7.1
+mdit-py-plugins==0.3.3
+mdurl==0.1.2
+multidict==6.0.4
+mutagen==1.46.0
+numpy==1.24.2
+openai==0.27.2
+orjson==3.8.8
+packaging==23.0
+pandas==1.5.3
+Pillow==9.4.0
+pycryptodomex==3.17
+pydantic==1.10.7
+pydub==0.25.1
+pyparsing==3.0.9
+pyrsistent==0.19.3
+python-dateutil==2.8.2
+python-multipart==0.0.6
+pytube==12.1.3
+pytz==2023.2
+PyYAML==6.0
+requests==2.28.2
+rfc3986==1.5.0
+semantic-version==2.10.0
+six==1.16.0
+sniffio==1.3.0
+starlette==0.26.1
+tensorflow==2.11.0
+tensorflow-intel==2.11.0
+toolz==0.12.0
+tqdm==4.65.0
+typing_extensions==4.5.0
+uc-micro-py==1.0.1
+urllib3==1.26.15
+uvicorn==0.21.1
+websockets==10.4
+wincertstore==0.2
+yarl==1.8.2
+youtube-dl==2021.12.17
+yt-dlp==2023.3.4
+zipp==3.15.0