chatbot implementation

README.md

@@ -11,9 +11,23 @@
 ## run Server
 
 - .\env\Scripts\activate
-- uvicorn index:app --reload
-
+- uvicorn main:app --reload
 
 ## Export Dependencies
+
 - .\env\Scripts\activate
-- pip freeze > requirements.txt
\ No newline at end of file
+- pip freeze > requirements.txt
+
+## Deploy to deta.sh
+
+- Install deta CLI
+
+  `iwr https://get.deta.dev/cli.ps1 -useb | iex`
+
+- Login to deta
+
+  `deta login`
+
+- Deploy
+
+  `deta new`

config/db.py

 from pymongo import MongoClient
-conn = MongoClient("mongodb+srv://skin-disease:skindis123@cluster0.bv8ia.mongodb.net/skinDiseaseDB?retryWrites=true&w=majority")
+conn = MongoClient("mongodb+srv://skin-disease:skindis123@cluster0.bv8ia.mongodb.net/skinDiseaseDB?retryWrites=true&w=majority", tls=True, tlsAllowInvalidCertificates=True)
+db = conn['skinDiseaseDB']
+disease_collection = db['disease']
 #mongodb+srv://skin-disease:skindis123@cluster0.bv8ia.mongodb.net/test
\ No newline at end of file

data.json

+{"23,2,8,9,11,12,68,15": "Acne Symptoms", "4,28,0,0,0,0,0,0": "Rosacea Symptoms", "53,32,6,21,39,0,0,0": "actinic keratosis symptoms", "84,81,82,72,85,0,0,0": "Atopic dermatitis", "37,42,55,27,0,0,0,0": "Bullous disease symptoms", "49,61,63,46,67,31,51,16": "Cellulitis Symptoms", "86,76,73,78,0,0,0,0": "Eczema Symptoms", "33,10,30,57,0,0,0,0": "Exanthems Symptoms", "56,47,34,70,7,5,0,0": "Alopecia symptoms", "75,71,45,13,0,0,0,0": "Herpes HPV symptoms", "79,77,74,83,0,0,0,0": "Light Disease Symptoms", "44,48,20,35,60,41,43,0": "lupus symptoms", "1,18,22,3,29,0,0,0": "melanoma symptoms", "66,69,19,25,14,59,0,0": "Nail Fungus symptoms", "52,40,61,17,24,0,0,0": "poison ivy symptoms", "50,58,26,38,65,62,0,0": "psoriasis symptoms", "36,80,54,64,0,0,0,0": "Scabies symptoms"}
\ No newline at end of file

index.py → main.py

 from fastapi import FastAPI
+from routes.disease import disease
 from routes.doctor import doctor
+from routes.prediction import prediction
 
 app = FastAPI()
+app.include_router(disease)
 app.include_router(doctor)
-
+app.include_router(prediction)

models/disease.py

+from pydantic import BaseModel
+from typing import Optional
+
+class Disease(BaseModel):
+    name: str
+    description: Optional[str]=""
\ No newline at end of file

requirements.txt

 asgiref==3.4.1

routes/chatbot.py

+from fastapi import APIRouter
+from transformers import AutoTokenizer, AutoModelWithLMHead, AutoConfig
+import torch
+from os import error
+from pydantic import BaseModel
+
+disease = APIRouter()
+
+class Message(BaseModel): 
+    msg: str
+
+@disease.post('/api/chatbot')
+async def create_disease(message: Message):
+    try:
+        tokenizerx = AutoTokenizer.from_pretrained(pretrained_model_name_or_path="t5-base")
+        model = AutoModelWithLMHead.from_pretrained("./chatbot/t5-bot/best_tfmr") 
+        text = message
+        device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+        preprocess_text = text.strip().replace("\n","")
+        tokenized_text = tokenizerx.encode(preprocess_text, return_tensors="pt").to(device)
+        model = model.to(device)
+        
+        outs = model.generate(
+                    tokenized_text,
+                    temperature=1.5,
+                    length_penalty=1.5,
+                    max_length=128,
+                    num_beams=1,
+                    early_stopping=True,
+                    no_repeat_ngram_size=2
+                )
+
+        dec = [tokenizerx.decode(ids) for ids in outs]
+        print("Predicted Answer: ", dec)
+
+        return {"success": True, "msg":dec}
+    except error: 
+        return {"success": False, "msg": error}

routes/disease.py

+from fastapi import APIRouter
+from config.db import conn
+from os import error
+
+from models.disease import Disease
+from schemas.disease import diseaseEntity, diseasesEntity
+
+disease = APIRouter()
+
+@disease.post('/api/disease')
+async def create_disease(disease: Disease):
+    try:
+        conn.skinDiseaseDB.disease.insert_one(dict(disease))
+        return {"success": True}
+    except error: 
+        return {"success": False, "msg": error}

routes/doctor.py

@@ -10,9 +10,15 @@ from schemas.availability import availabilityEntity, availabilitiesEntity
 from fastapi import APIRouter
 from models.doctor import Doctor
 from config.db import conn
+from pydantic import BaseModel
+
 
 doctor = APIRouter()
 
+
+class Availabilities(BaseModel): 
+    availabilities: List[Availability]
+
 @doctor.get('/api/doctors')
 async def find_all_doctors():
     return doctorsEntity(conn.skinDiseaseDB.doctor.find())
@@ -52,14 +58,10 @@ async def delete_doctor(id):
 
 
 @doctor.put('/api/availability/{id}')
-async def update_availability(id, availabilities: List[Availability]):
-    try:
-        available_dic = dict()
-        for available in availabilities:
-            print(available)
-        
-        #conn.skinDiseaseDB.doctor.find_one_and_update({"id":id},{"$set": {'isAvailable': isAvailable}})
-        return availabilities
+async def update_availability(id, availabilities: Availabilities):
+    try: 
+        conn.skinDiseaseDB.doctor.find_one_and_update({"serviceID": id}, {"$set": {'availability': availabilities }}) 
+        return {"success": True, } 
     except error: 
         return {"success": False, "msg": error}
 

routes/prediction.py

+from fastapi import APIRouter, File, UploadFile
+from pydantic import BaseModel
+# import tensorflow as tf
+# from tensorflow import keras 
+from typing import List
+from os import error
+# import pandas as pd
+# import numpy as np
+import json
+#from keras.preprocessing import image 
+from config.db import disease_collection
+from schemas.disease import diseasesEntity
+
+
+prediction = APIRouter()
+
+class Symptoms(BaseModel):
+    symptoms: List[int] 
+
+batch_size = 32
+img_height = 180
+img_width = 180
+class_names = ['Acne and Rosacea Photos', 'Actinic Keratosis Basal Cell Carcinoma and other Malignant Lesions', 'Atopic Dermatitis Photos', 'Bullous Disease Photos', 'Cellulitis Impetigo and other Bacterial Infections', 'Eczema Photos', 'Exanthems and Drug Eruptions', 'Hair Loss Photos Alopecia and other Hair Diseases', 'Herpes HPV and other STDs Photos', 'Light Diseases and Disorders of Pigmentation', 'Lupus and other Connective Tissue diseases', 'Melanoma Skin Cancer Nevi and Moles', 'Nail Fungus and other Nail Disease', 'Poison Ivy Photos and other Contact Dermatitis', 'Psoriasis pictures Lichen Planus and related diseases', 'Scabies Lyme Disease and other Infestations and Bites', 'Seborrheic Keratoses and other Benign Tumors', 'Systemic Disease', 'Tinea Ringworm Candidiasis and other Fungal Infections', 'Urticaria Hives', 'Vascular Tumors', 'Vasculitis Photos', 'Warts Molluscum and other Viral Infections']
+
+
+def load_data_json(path=''):
+  with open(path, 'r') as fp:
+    data = json.load(fp)
+  fp.close()
+  return data
+
+def is_image(filename: str)->bool:
+  valid = (".png", ".jpg", "jpeg")
+  return filename.endswith(valid)
+
+def read_imagefile( image:UploadFile ):
+  if is_image(image.filename):
+    directory = './temp_images/temp_image.jpg'
+    with open(directory, 'wb+') as image_upload:
+      image_upload.write(image.file.read())
+    return directory
+  return None
+
+
+# PREDICT DISEASE FROM SYMPTOMSES
+@prediction.post('/api/predictions/symptoms')
+async def predict_symptoms(symptoms: Symptoms):
+    try:
+        data = load_data_json('./data.json')
+        threshold = 0.5
+        input = set(symptoms.symptoms)
+        output = []
+        for key in data.keys():
+            numbers = [int(i) for i in key.split(',') if int(i) != 0]
+            lenght = len(numbers)
+            score = len(list(input&set(numbers)))/lenght
+            if score>=threshold:
+                output.append(data.get(key))
+        
+        diseases = diseasesEntity(disease_collection.find({"name": {"$in": output}}))  
+                
+        return {"success": True, "result": diseases}  
+        
+    except error: 
+        return {"success": False, "msg": error}
+
+
+# PREDICT DISEASE FROM PICTURES
+@prediction.post('/api/predictions/picture')
+async def predict_symptoms(req_file: UploadFile = File(...)):
+    try:
+      # img = read_imagefile(req_file)
+
+      # if img is None:
+      #   return {"success": False, "msg": 'Not a valid data format'}
+
+      # model_path = './xception'
+      # image_path = "./temp_images/temp_image.jpg"
+
+      # #load model
+      # model = keras.models.load_model(model_path)
+
+      # #load data
+      # load_img = image.load_img(image_path, target_size=(img_height, img_width))
+      # img_array = image.img_to_array(load_img)
+      # img_array = tf.expand_dims(img_array, 0)
+
+      # #predict
+      # predictions = model.predict(img_array, batch_size)
+      # score = tf.nn.softmax(predictions[0])
+
+      # #return
+      # prediction = class_names[np.argmax(score)]
+
+      return {"success": True, "result": req_file.filename}
+      
+    except error: 
+      return {"success": False, "msg": error}
+
+
+
+

schemas/disease.py

+def diseaseEntity(item) -> dict: 
+    return { 
+        "name": item["name"],
+        "description": item["description"],
+    }
+
+def diseasesEntity(entity) -> list:
+    return [diseaseEntity(item) for item in entity]
\ No newline at end of file

schemas/doctor.py

@@ -20,4 +20,4 @@ def doctorGetEntity(item) -> dict:
     }
 
 def doctorsEntity(entity) -> list:
-    return [doctorGetEntity(item) for item in entity]
\ No newline at end of file
+    return [doctorGetEntity(item) for item in entity]