Duplicate from epochs-demos/product-recommendor

.gitattributes

@@ -0,0 +1,34 @@
+*.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
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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

README.md

@@ -0,0 +1,13 @@
+---
+title: Product Recommendor
+emoji: 👁
+colorFrom: gray
+colorTo: blue
+sdk: gradio
+sdk_version: 3.29.0
+app_file: app.py
+pinned: false
+duplicated_from: epochs-demos/product-recommendor
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,336 @@
+#Importing all the necessary needed  libraries
+import torch
+import requests
+import numpy as np
+import pandas as pd
+import gradio as gr
+from io import BytesIO
+from PIL import Image as PILIMAGE
+from IPython.display import Image
+from IPython.core.display import HTML
+from transformers import CLIPProcessor, CLIPModel, CLIPTokenizer
+from sentence_transformers import SentenceTransformer, util
+import os
+import json
+import requests
+import langchain
+from tqdm import tqdm
+from langchain.text_splitter import CharacterTextSplitter
+images = [] 
+prompt_templates = {"DefaultChatGPT": ""}
+# Streaming endpoint
+API_URL = "https://api.openai.com/v1/chat/completions"  # os.getenv("API_URL") + "/generate_stream"
+convo_id = 'default'
+#5c72c157a8fd54357bd13112cd71952a
+import time
+images1= pd.read_csv("./images.csv")
+
+openai_api_key='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'
+m=0
+style1= pd.read_csv('./stylesu.csv')
+feature_info= list(style1.columns)
+feature_info = ' '.join([str(elem) for elem in feature_info])
+info= style1.values.tolist()
+final_info=''
+for i in info:
+    li=''
+    li=' '.join([str(elem) for elem in i])
+    final_info += li+'\n'
+
+
+
+def on_prompt_template_change(prompt_template):
+    if not isinstance(prompt_template, str): return
+    if prompt_template:
+        return prompt_templates[prompt_template]
+    else:
+        ''
+
+def get_empty_state():
+    return {"total_tokens": 0, "messages": []}
+
+def get_prompt_templates():
+    with open('./prompts.json','r',encoding='utf8') as fp:
+        json_data = json.load(fp)
+        for data in json_data:
+            act = data['act']
+            prompt = data['prompt']
+            prompt_templates[act] = prompt
+        # reader = csv.reader(csv_file)
+        # next(reader)  # skip the header row
+        # for row in reader:
+        #     if len(row) >= 2:
+        #         act = row[0].strip('"')
+        #         prompt = row[1].strip('"')
+        #         prompt_templates[act] = prompt
+
+        choices = list(prompt_templates.keys())
+        choices = choices[:1] + sorted(choices[1:])
+        return gr.update(value=choices[0], choices=choices)
+
+
+
+def run(pr=gr.Progress(track_tqdm=True)):
+    #if(chat_counter==0):
+    message_prompt=[]
+    x=len(final_info)
+    print(x/2000)
+    for i in range(0,x,2000): #final_texts:
+        message_prompt.append(final_info[i:i+2000]+" Remember this along with previous prompts as it makes up the csv file")
+            #//there
+    prompt_template = "I want you to act as a Product recommender and read the CSV file I will provide you. I need you to thoroughly review the CSV file and give recommendations based on the input afterward. You should recommend me the product by displaying its id, and description. The csv features are:" +feature_info+ "The csv information is as follows:"
+    payload = {
+                    "model": "gpt-3.5-turbo",
+                    "messages": [{"role":"system", "content":prompt_template}],
+                    "temperature": 0.1,
+                    "top_p": 1.0,
+                    "n": 1,
+                    "stream": True,
+                    "presence_penalty": 0,
+                    "frequency_penalty": 0,
+                }
+            
+    headers = {
+                    "Content-Type": "application/json",
+                    "Authorization": f"Bearer {openai_api_key}"
+                }
+    response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+          
+    for i in pr.tqdm(message_prompt):
+        payload = {
+                    "model": "gpt-3.5-turbo",
+                    "messages": [{"role":"system", "content":i}],
+                    "temperature": 0.1,
+                    "top_p": 1.0,
+                    "n": 1,
+                    "stream": True,
+                    "presence_penalty": 0,
+                    "frequency_penalty": 0, }
+        response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+        time.sleep(0.01)
+        pr(1/2210)
+
+    print("completed")
+    
+def predict(inputs, prompt_template, temperature, openai_api_key, chat_counter, context_length, chatbot=[],
+            history=[]):
+
+ # # repetition_penalty, top_k
+    if inputs==None:
+        inputs = ''
+    prompt_template = "I want you to act as a Product recommender and read the CSV file I will provide you. I need you to thoroughly review the CSV file and give recommendations based on the input afterward. You should recommend me the product by displaying its id, and description. The csv features are:" +feature_info+ "The csv information is as follows:"
+        
+    headers = {
+                    "Content-Type": "application/json",
+                    "Authorization": f"Bearer {openai_api_key}"
+                }    
+    payload = {
+        "model": "gpt-3.5-turbo",
+        "messages": [{"role": "user", "content": f"{inputs}"}],
+        "temperature": 0.1,
+        "top_p": 1.0,
+        "n": 1,
+        "stream": True,
+        "presence_penalty": 0,
+        "frequency_penalty": 0,
+    }
+    
+        
+        
+                # print(f"chat_counter - {chat_counter}")
+    if chat_counter != 0:
+        messages = []
+        # print(chatbot)
+        # print(chatbot[-context_length:])
+        # print(context_length)
+        for data in chatbot[-context_length:]:
+            temp1 = {}
+            temp1["role"] = "user"
+            temp1["content"] = data[0]
+            temp2 = {}
+            temp2["role"] = "assistant"
+            temp2["content"] = data[1]
+            messages.append(temp1)
+            messages.append(temp2)
+        temp3 = {}
+        temp3["role"] = "user"
+        temp3["content"] = inputs
+        messages.append(temp3)
+        # print(messages)
+        # messages
+        payload = {
+            "model": "gpt-3.5-turbo",
+            "messages": [{"role": "system", "content": prompt_template}]+messages,  # [{"role": "user", "content": f"{inputs}"}],
+            "temperature": temperature,  # 1.0,
+            "n": 1,
+            "stream": True,
+            "presence_penalty": 0,
+            "frequency_penalty": 0,
+        }
+
+
+
+    history.append(inputs)
+    # print(f"payload is - {payload}")
+    # make a POST request to the API endpoint using the requests.post method, passing in stream=True
+    # print('payload',payload)
+    response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+
+    # print('response', response)
+    # print('content',response.content)
+    # print('text', response.text)
+    if response.status_code != 200:
+        try:
+            payload['id'] = response.content['id']
+            response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+            if response.status_code != 200:
+                payload['id'] = response.content['id']
+                response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+        except:
+            pass
+
+    # print('status_code', response.status_code)
+    # response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+    token_counter = 0
+    partial_words = ""
+    counter = 0
+    if response.status_code==200:
+        chat_counter += 1
+        # print('chunk')
+        for chunk in response.iter_lines():
+            # Skipping first chunk
+            if counter == 0:
+                counter += 1
+                continue
+            # check whether each line is non-empty
+            chunk = chunk.decode("utf-8")[6:]
+            if chunk:
+                # print(chunk)
+                if chunk=='[DONE]':
+                    break
+                resp: dict = json.loads(chunk)
+                choices = resp.get("choices")
+                if not choices:
+                    continue
+                delta = choices[0].get("delta")
+                if not delta:
+                    continue
+                # decode each line as response data is in bytes
+                if len(chunk) > 12 and "content" in resp['choices'][0]['delta']:
+                    # if len(json.loads(chunk.decode()[6:])['choices'][0]["delta"]) == 0:
+                    #  break
+                    partial_words = partial_words + resp['choices'][0]["delta"]["content"]
+                    # print(partial_words)
+                    if token_counter == 0:
+                        history.append(" " + partial_words)
+                    else:
+                        history[-1] = partial_words
+                    chat = [(history[i], history[i + 1]) for i in
+                            range(0, len(history) - 1, 2)]  # convert to tuples of list
+                    # print(chat)
+                    token_counter += 1
+                    yield chat, history, chat_counter  # resembles {chatbot: chat, state: history}
+    else:
+        chat = [(history[i], history[i + 1]) for i in
+                range(0, len(history) - 1, 2)]  # convert to tuples of list
+        chat.append((inputs, "OpenAI Network Error. please try again"))
+        token_counter += 1
+        yield chat, history, chat_counter  # resembles {chatbot: chat, state: history}
+       
+
+
+
+def reset_textbox():
+    return gr.update(value='')
+
+def clear_conversation(chatbot):
+    return gr.update(value=None, visible=True), [], [], gr.update(value=0)
+
+
+   
+def galleryim():
+    
+    count=0
+    for i in images1['filename']:
+        count+=1
+        if count==50:
+             break
+        photo_data = images1[images1["filename"] == i].iloc[0]
+        response = requests.get(photo_data["link"] )
+        try:
+            img = PILIMAGE.open(BytesIO(response.content))
+        except:
+            print("File not found")
+        else:
+            images.append(img)
+    return images
+
+title = """<h1 align="center">ChatGPTDatasetSearch</h1>"""
+description = """Language models can be conditioned to act like dialogue agents through a conversational prompt that typically takes the form:
+```
+User: <utterance>
+Assistant: <utterance>
+User: <utterance>
+Assistant: <utterance>
+...
+```
+In this app, you can explore the outputs of a gpt-3.5-turbo LLM.
+"""
+with gr.Blocks(css="""#col_container {width: 800px; margin-left: auto; margin-right: auto;}
+                #chatbot {height: 500px; overflow: auto;}
+                #inputs {font-size: 20px;}
+                #prompt_template_preview {padding: 1em; border-width: 1px; border-style: solid; border-color: #e0e0e0; border-radius: 4px;}""") as demo:
+    gr.HTML(title)
+    with gr.Column(variant="panel"):
+        
+        gr.HTML(  """<b><center><h1>1001Epochs</h1></center></b>
+                    <p><center>TOP THREE images that best match the search query provided by the user</center></p>
+                    """)
+    
+    with gr.Row():
+        with gr.Column(scale=0.50):
+            gallery = gr.Gallery( value=galleryim(),
+            label="Generated images", show_label=False, elem_id="gallery",every=60).style(columns=5, container=True)
+       
+        with gr.Column(elem_id="col_container"):
+            
+            openai_api_key = gr.Textbox(type='password', label="Enter API Key",placeholder="sk-xxxxxxxx")
+            button1=gr.Button("feed the csv into model")
+            button1.click(run, show_progress=True)
+            chatbot = gr.Chatbot(elem_id='chatbot')  # c
+            inputs = gr.Textbox(show_label=False, placeholder="Enter Content",elem_id="inputs",value='')  # t
+            state = gr.State([])  # s
+            # state = gr.State(get_empty_state())
+            b1 = gr.Button("Submit")
+            btn_clear_conversation = gr.Button("🔃 New Conversation")
+    
+            # inputs, top_p, temperature, top_k, repetition_penalty
+            with gr.Accordion("Advanced settings", open=False,):
+                context_length = gr.Slider(minimum=1, maximum=6, value=2, step=1, label="Dialogue Length",
+                                           info="Associate the previous rounds of dialogues, the higher the value, the more tokens will be consumed")
+                temperature = gr.Slider(minimum=0, maximum=2.0, value=0.7, step=0.1, label="Temperature",
+                                        info="The higher the value, the stronger the creativity")
+                prompt_template = gr.Dropdown(label="Choose robot type",
+                                              choices=list(prompt_templates.keys()),visible=False)
+                prompt_template_preview = gr.Markdown(elem_id="prompt_template_preview",visible=False)
+                # top_k = gr.Slider( minimum=1, maximum=50, value=4, step=1, interactive=True, label="Top-k",)
+                # repetition_penalty = gr.Slider( minimum=0.1, maximum=3.0, value=1.03, step=0.01, interactive=True, label="Repetition Penalty", )
+                chat_counter = gr.Number(value=0, visible=False, precision=0)
+
+    inputs.submit(predict, [inputs, prompt_template, temperature, openai_api_key, chat_counter, context_length, chatbot, state],
+                  [chatbot, state, chat_counter], )
+    b1.click(predict, [inputs, prompt_template, temperature, openai_api_key, chat_counter, context_length, chatbot, state],
+             [chatbot, state, chat_counter], )
+    b1.click(reset_textbox, [], [inputs])
+
+    btn_clear_conversation.click(clear_conversation, [], [inputs, chatbot, state, chat_counter])
+
+    inputs.submit(reset_textbox, [], [inputs])
+    prompt_template.change(on_prompt_template_change, inputs=[prompt_template], outputs=[prompt_template_preview])
+    demo.load(get_prompt_templates, inputs=None, outputs=[prompt_template], queur=False)
+
+    # gr.Markdown(description)
+    demo.queue(concurrency_count=10)
+    demo.launch(debug=True)
+
+

prompts.json

@@ -0,0 +1,7 @@
+[
+    {
+      "act": "Act as product Recommendation system",
+      "prompt": "I want you to act as a Product recommender and read the CSV file I will provide you. I need you to thoroughly review the CSV file and give recommendations based on the input afterward. You should recommend me the product by displaying its id,description.\n"
+    },
+    
+  ]

requirements.txt

@@ -0,0 +1,9 @@
+sentence-transformers==2.2.1
+transformers
+torch
+numpy
+tiktoken
+
+ftfy
+langchain
+IPython