Init

.gitattributes

@@ -1,4 +1,4 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
+preprocess_txt_split*.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+sentences.csv filter=lfs diff=lfs merge=lfs -text
+preprocess_df_count_word* filter=lfs diff=lfs merge=lfs -text 
+preprocess_txt_split* filter=lfs diff=lfs merge=lfs -text
+

.gitignore

@@ -0,0 +1,7 @@
+# DotEnv configuration
+.env
+#
+*.pyc
+
+/tabs/__pycache__
+/__pycache__

README.md

@@ -1,13 +1,12 @@
 ---
-title: ChatGPT Playground
-emoji: 📈
-colorFrom: purple
-colorTo: purple
+title: Playgrount ChatGPT
+emoji: 🏆
+colorFrom: red
+colorTo: red
 sdk: streamlit
-sdk_version: 1.41.1
+sdk_version: 1.38.0
 app_file: app.py
 pinned: false
-short_description: .. to see the token generation (Model used=davinci-002)
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,111 @@
+import streamlit as st # type: ignore
+import os.path
+from collections import OrderedDict
+from streamlit_option_menu import option_menu # type: ignore
+# Define TITLE, TEAM_MEMBERS and PROMOTION values, in config.py.
+import config
+import os
+import openai
+from dotenv import load_dotenv
+from translate_app import tr
+
+# Initialize a session state variable that tracks the sidebar state (either 'expanded' or 'collapsed').
+if 'sidebar_state' not in st.session_state:
+    st.session_state.sidebar_state = 'expanded'
+else:
+    st.session_state.sidebar_state = 'auto'
+
+st.set_page_config (
+    page_title=config.TITLE,
+    page_icon= "assets/favicon.png",
+    initial_sidebar_state=st.session_state.sidebar_state
+)
+
+# Si l'application tourne localement, session_state.Cloud == 0 
+# Si elle tourne sur le Cloud de Hugging Face, ==1
+st.session_state.Cloud = 0
+st.session_state.DataPath = "data"
+st.session_state.ImagePath = "images"
+st.session_state.reCalcule = False
+
+# Define the root folders depending on local/cloud run
+# thisfile = os.path.abspath(__file__)
+# if ('/' in thisfile): 
+#     os.chdir(os.path.dirname(thisfile))
+
+# Nécessaire pour la version windows 11
+if st.session_state.Cloud == 0: 
+    os.environ['PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION'] = 'python'
+    
+
+# Tabs in the ./tabs folder, imported here.
+from tabs import playground
+
+
+with open("style.css", "r") as f:
+    style = f.read()
+
+st.markdown(f"<style>{style}</style>", unsafe_allow_html=True)
+
+
+# Add tab in this ordered dict by
+# passing the name in the sidebar as key and the imported tab
+# as value as follow :
+TABS = OrderedDict(
+    [
+        (tr(playground.sidebar_name), playground),
+        # (tr(sentence_similarity_tab.sidebar_name), sentence_similarity_tab),
+        # (tr(speech2text_tab.sidebar_name), speech2text_tab),
+        # (tr(chatbot_tab.sidebar_name), chatbot_tab),
+    ]
+)
+
+# Utilisation du module deep_translator
+lang_tgt = ['fr', 'en', 'af', 'ak', 'sq', 'de', 'am', 'en', 'ar', 'hy', 'as', 'ay', 'az', 'bm', 'eu', 'bn', 'bho', 'be', 'my', 'bs', 'bg', 'ca', 'ceb', 'ny', 'zh-CN', 'zh-TW', 'si', 'ko', 'co', 'ht', 'hr', 'da', 'doi', 'gd', 'es', 'eo', 'et', 'ee', 'fi', 'fr', 'fy', 'gl', 'cy', 'lg', 'ka', 'el', 'gn', 'gu', 'ha', 'haw', 'iw', 'hi', 'hmn', 'hu', 'ig', 'ilo', 'id', 'ga', 'is', 'it', 'ja', 'jw', 'kn', 'kk', 'km', 'rw', 'ky', 'gom', 'kri', 'ku', 'ckb', 'lo', 'la', 'lv', 'ln', 'lt', 'lb', 'mk', 'mai', 'ms', 'ml', 'dv', 'mg', 'mt', 'mi', 'mr', 'mni-Mtei', 'lus', 'mn', 'nl', 'ne', 'no','or', 'om', 'ug', 'ur', 'uz', 'ps', 'pa', 'fa', 'pl', 'pt', 'qu', 'ro', 'ru', 'sm', 'sa', 'nso', 'sr', 'sn', 'sd', 'sk', 'sl', 'so', 'st', 'su', 'sv', 'sw', 'tg', 'tl', 'ta', 'tt', 'cs', 'te', 'th', 'ti', 'ts', 'tr', 'tk', 'uk', 'vi', 'xh', 'yi', 'yo', 'zu']
+label_lang = ['Français', 'Anglais / English','Afrikaans','Akan','Albanais','Allemand / Deutsch','Amharique','Anglais','Arabe','Arménien','Assamais','Aymara','Azéri','Bambara','Basque','Bengali','Bhojpuri','Biélorusse','Birman','Bosnien','Bulgare','Catalan','Cebuano','Chichewa','Chinois (simplifié)','Chinois (traditionnel)','Cingalais','Coréen','Corse','Créole haïtien','Croate','Danois','Dogri','Écossais','Espagnol / Español','Espéranto','Estonien','Ewe','Finnois','Français','Frisonoccidental','Galicien','Gallois','Ganda','Géorgien','Grec moderne','Guarani','Gujarati','Haoussa','Hawaïen','Hébreu','Hindi','Hmong','Hongrois','Igbo','Ilocano','Indonésien','Irlandais','Islandais','Italien / Italiano','Japonais','Javanais','Kannada','Kazakh','Khmer','Kinyarwanda','Kirghiz','Konkani','Krio','Kurde','Kurde (Sorani)','Lao','Latin','Letton','Lingala','Lituanien','Luxembourgeois','Macédonien','Maithili','Malais','Malayalam','Maldivien','Malgache','Maltais','Maori de Nouvelle-Zélande','Marathi','Meiteilon (Manipuri)','Mizo','Mongol','Néerlandais / Nederlands','Népalais','Norvégien','Oriya','Oromo','Ouïghour','Ourdou','Ouzbek','Pachto','Pendjabi','Persan','Polonais','Portugais','Quechua','Roumain','Russe','Samoan','Sanskrit','Sepedi','Serbe','Shona','Sindhi','Slovaque','Slovène','Somali','Sotho du Sud','Soundanais','Suédois','Swahili','Tadjik','Tagalog','Tamoul','Tatar','Tchèque','Télougou','Thaï','Tigrigna','Tsonga','Turc','Turkmène','Ukrainien','Vietnamien','Xhosa','Yiddish','Yoruba','Zulu']
+
+@st.cache_data        
+def find_lang_label(lang_sel):
+    global lang_tgt, label_lang
+    return label_lang[lang_tgt.index(lang_sel)]
+
+def run():
+
+    st.sidebar.image(
+        "assets/Logo_TPA.png",
+        width=270,
+    )
+    with st.sidebar:
+        tab_name = option_menu(None, list(TABS.keys()),
+                               # icons=['house', 'bi-binoculars', 'bi bi-graph-up', 'bi-chat-right-text','bi-book', 'bi-body-text'], menu_icon="cast", default_index=0,
+                               icons=['house', 'binoculars', 'graph-up', 'search','book', 'chat-right-text','controller'], menu_icon="cast", default_index=0,
+                               styles={"container": {"padding": "0!important","background-color": "#10b8dd", "border-radius": "0!important"},
+                                       "nav-link": {"font-size": "1rem", "text-align": "left", "margin":"0em", "padding": "0em",
+                                                    "padding-left": "0.2em", "--hover-color": "#eee", "font-weight": "400",
+                                                    "font-family": "Source Sans Pro, sans-serif"}
+                                        })
+    st.sidebar.markdown("---")
+    st.sidebar.markdown(f"## {tr(config.PROMOTION)}")
+
+    st.sidebar.markdown("### "+tr("Auteur:"), unsafe_allow_html=True)
+    for member in config.TEAM_MEMBERS:
+        st.sidebar.markdown(member.sidebar_markdown(), unsafe_allow_html=True)
+
+    with st.sidebar:
+        st.write("")
+        if ('OPENAI_API_KEY' not in st.session_state):
+            # Set OpenAI API key
+            st.sidebar.subheader("OpenAI API Key")
+            openai_api_key = st.sidebar.text_input(tr("Saisissez votre Clé API OpenAI:"), type='password')
+            if openai_api_key:
+                os.environ['OPENAI_API_KEY'] = openai_api_key
+                st.session_state['OPENAI_API_KEY'] = openai_api_key
+                st.sidebar.success("OpenAI API Key set successfully.")
+    with st.sidebar:
+        st.selectbox("langue:",lang_tgt, format_func = find_lang_label, key="Language", label_visibility="hidden")
+
+    tab = TABS[tab_name]
+    tab.run()
+
+if __name__ == "__main__":
+    run()

config.py

@@ -0,0 +1,23 @@
+"""
+
+Config file for Streamlit App
+
+"""
+
+from member import Member
+
+
+TITLE = "ChatGPT"
+
+TEAM_MEMBERS = [
+ 
+    Member(
+    name="Olivier Renouard",
+    linkedin_url="https://www.linkedin.com/in/olivier-renouard/",
+    github_url="https://github.com/Demosthene-OR",
+    )
+    
+
+]
+
+PROMOTION = "Playground ChatGPT - Octobre 2024"

member.py

@@ -0,0 +1,18 @@
+class Member:
+    def __init__(
+        self, name: str, linkedin_url: str = None, github_url: str = None
+    ) -> None:
+        self.name = name
+        self.linkedin_url = linkedin_url
+        self.github_url = github_url
+
+    def sidebar_markdown(self):
+
+        markdown = f'<b style="display: inline-block; vertical-align: middle; height: 100%">{self.name}</b>'
+        if self.linkedin_url is not None:
+            markdown += f' <a href={self.linkedin_url} target="_blank"><img src="https://dst-studio-template.s3.eu-west-3.amazonaws.com/linkedin-logo-black.png" alt="linkedin" width="25" style="vertical-align: middle; margin-left: 5px"/></a> '
+
+        if self.github_url is not None:
+            markdown += f' <a href={self.github_url} target="_blank"><img src="https://dst-studio-template.s3.eu-west-3.amazonaws.com/github-logo.png" alt="github" width="20" style="vertical-align: middle; margin-left: 5px"/></a> '
+        print(markdown)
+        return markdown

requirements.txt

@@ -0,0 +1,9 @@
+streamlit==1.34.0
+pandas==2.2.2
+ipython==8.27.0
+numpy==1.26.4
+https://files.pythonhosted.org/packages/cc/58/96aff0e5cb8b59c06232ea7e249ed902d04ec89f52636f5be06ceb0855fe/extra_streamlit_components-0.1.60-py3-none-any.whl
+streamlit-option-menu==0.3.12
+deep-translator==1.11.4
+openai==0.27.8
+python-dotenv==1.0.1

style.css

@@ -0,0 +1,138 @@
+h1 {
+  padding-top: 0rem;
+  padding-bottom: 0rem;
+  margin-top:6px;
+}
+h2 {
+  padding-top: 0.75rem;
+  padding-bottom: 0.5rem;
+}
+
+h3 {
+  padding-bottom: 0.0rem;
+}
+
+
+/* La ligne suivante est nécessaire à cause du module streamlit_option_menu qui "casse" les CSS suivants */
+@media (prefers-color-scheme: dark) {
+  .st-cc {
+    color: #fff!important; /* Couleur du texte en mode sombre */
+  }
+	.st-cg:hover {
+		color: rgb(255, 75, 75)!important; /* Couleur du texte en mode sombre */
+  }
+  section[data-testid="stSidebar"] .stSelectbox .st-cc {
+    color: rgb(255, 75, 75)!important;
+    font-weight: bold;
+  }
+}
+
+p {
+    margin-bottom:0.1rem;
+}
+
+code {
+  color: #1ec3bc;
+}
+
+#MainMenu {
+  display: none;
+}
+
+div[data-testid="stDecoration"] {
+  display: none;
+}
+
+footer {
+  display: none;
+}
+
+/* Radio buttons */
+
+.st-cc {
+  color: black;
+  font-weight: 500;
+}
+
+/* Sidebar */
+
+.css-1544g2n {
+  padding-top: 1rem;
+}
+
+.css-10oheav {
+  padding-top: 3rem;
+}
+
+.css-ue6h4q {
+  min-height: 0.5rem;
+}
+
+section[data-testid="stSidebar"] > div {
+  background-color: #10b8dd;
+  padding-top: 1rem;
+  padding-left: 0.5rem;
+}
+
+section[data-testid="stSidebar"] button[title="View fullscreen"] {
+  display: none;
+}
+
+section[data-testid="stSidebar"] button[kind="icon"] {
+  display: none;
+}
+
+section[data-testid="stSidebar"] .st-bk {
+  background-color: #10b8dd;
+}
+
+section[data-testid="stSidebar"] .st-c0 {
+/*  background-color: #10b8dd; */
+}
+
+section[data-testid="stSidebar"] hr {
+  margin-top: 30px;
+  border-color: white;
+  width: 50px;
+}
+
+section[data-testid="stSidebar"] h2 {
+  color: white;
+}
+
+section[data-testid="stSidebar"] .stSelectbox .st-bk {
+  background-color: #a0d3de;
+}
+
+section[data-testid="stSidebar"] .stSelectbox .st-cc {
+  color: rgb(255, 75, 75);
+  font-weight: bold;
+}
+
+/* Images */
+
+button[title="View fullscreen"] {
+  display: none;
+}
+
+/* hr */
+
+hr {
+  width: 700px;
+  border-width: 5px;
+  border-color: #10b8dd;
+  margin-top: 0px;
+  margin-bottom: 1em;
+  max-width: 100%;
+}
+
+/* First Page */
+
+section[tabindex="0"] .block-container {
+  padding-top: 0px;
+  padding-bottom: 0px;
+}
+
+.st-emotion-cache-12fmjuu {
+  height: 0rem;
+}

tabs/playground.py

@@ -0,0 +1,115 @@
+import streamlit as st # type: ignore
+from translate_app import tr
+import numpy as np
+import openai
+from dotenv import load_dotenv
+import os
+import warnings
+warnings.filterwarnings('ignore')
+
+if st.session_state.Cloud != 0:
+    load_dotenv()
+os.getenv("OPENAI_API_KEY")
+openai.api_key = os.environ["OPENAI_API_KEY"]
+
+title = "Playground ChatGPT"
+sidebar_name = "Playground"
+
+def display_proba_next_token(prompt,temperature,max_token):
+
+    # Définir un ou plusieurs motifs d'arrêt
+    stop_sequences = [".","\n", "FIN", "Merci"]  # Ajoutez d'autres motifs selon vos besoins
+    response = openai.Completion.create(
+        engine="davinci-002",
+        prompt=prompt,
+        max_tokens=max_token,       # Limite à un seul token pour voir uniquement la probabilité du prochain
+        logprobs=5,                 # Obtenir les 5 tokens les plus probables
+        temperature=temperature,     # Contrôle la créativité, 0 pour des réponses plus prédictibles
+        stop=stop_sequences 
+    )
+
+    # Mettre logprobs dans un dictionnaire
+    if response['choices'][0]['text'] !="":
+        logprobs = response['choices'][0]['logprobs']['top_logprobs'][0]
+        logprobs_dict = dict(logprobs)
+
+        # Convertir logprobs en probabilités
+        probabilities = {key: np.exp(value) for key, value in logprobs_dict.items()}
+
+        # Afficher le texte de la réponse
+        st.write("Token suivant : ->"+list(probabilities.keys())[0]+"<-")  # Afficher le texte de la réponse
+        # st.write("Texte de la réponse : ->"+response['choices'][0]['text'][0]+"<-")  # Afficher le texte de la réponse
+        st.write("")
+
+        # Afficher les probabilités pour le prochain token
+        for token, proba in probabilities.items():
+            st.write(f"Token: ->{token}<- ; proba: {proba:.2f}")
+
+        # Afficher le texte de la réponse complete
+        st.write("")
+        try:
+            text = response['choices'][0]['text']
+            st.write("Texte de la réponse complète: ->" + str(text) + "<-")
+        except KeyError:
+            st.error("La réponse ne contient pas de texte supplémentaire.")
+    else:
+        st.error("La réponse ne contient pas de texte supplémentaire.")
+    # st.write("Texte de la réponse complete: ->"+str(response['choices'][0]['text'])+"<-")  # Afficher le texte de la réponse
+    # st.write("")
+    # st.write(response)
+
+
+def run():
+    global temperature, max_token
+    
+    st.write("")
+    st.write("")
+    # TODO: choose between one of these GIFs
+    # st.image("https://dst-studio-template.s3.eu-west-3.amazonaws.com/1.gif")
+    # st.image("https://dst-studio-template.s3.eu-west-3.amazonaws.com/2.gif")
+    # st.image("https://dst-studio-template.s3.eu-west-3.amazonaws.com/3.gif")
+    # st.image("assets/tough-communication.gif",use_column_width=True)
+    
+    
+    # if st.session_state.Cloud == 0: 
+    #     st.image("assets/miss-honey-glasses-off.gif",use_column_width=True)
+    # else:
+    #     st.image("https://media.tenor.com/pfOeAfytY98AAAAC/miss-honey-glasses-off.gif",use_column_width=True)
+        
+    st.title(tr(title))
+    st.markdown('''
+                ---
+                ''')
+    st.header("**"+tr("Prédiction du token suivant")+"**")
+    st.markdown(tr(
+        """
+        Cet espace a pour objectif d'observer la génération de token par ChatGPT'
+        """)
+    , unsafe_allow_html=True)
+
+    try:
+        col1, col2 = st.columns([3, 1])
+        with col2:
+            temperature = st.slider(
+                label=tr("Temperature"),
+                min_value=0.0,
+                max_value=1.0,
+                value=0.0  # Valeur initiale
+                )
+            max_token = st.slider(
+                label=tr("Max tokens"),
+                min_value=1,
+                max_value=500,
+                value=100 # Valeur initiale
+                ) 
+        with col1:
+            prompt = st.text_area(label=tr("Prompt:"), value="La souris est mangée par le",height=100)
+            display_proba_next_token(prompt,temperature,max_token)
+    except Exception as e:
+        st.error(f"An unexpected error occurred: {e}")
+        
+
+    
+    
+
+

translate_app.py

@@ -0,0 +1,27 @@
+import streamlit as st
+# from translate import Translator
+from deep_translator import GoogleTranslator
+
+@st.cache_data(ttl="2d", show_spinner=False)
+def trad(message,l):
+    try:
+        # Utilisation du module translate
+        # translator = Translator(to_lang=l , from_lang="fr")
+        # translation = translator.translate(message)
+
+        # Utilisation du module deep_translator 
+        translation = GoogleTranslator(source='fr', target=l).translate(message.replace("  \n","§§§"))
+        translation = translation.replace("§§§","  \n") # .replace("  ","<br>")
+
+        return translation
+    except:
+        return "Problème de traduction.."
+    
+def tr(message):
+    if 'Language' not in st.session_state: l = 'fr'
+    else: l= st.session_state['Language']
+    if l == 'fr': return message
+    else: message = message.replace(":red[**","").replace("**]","")
+    return trad(message,l)
+
+