Spaces:
Running
Running
File size: 11,457 Bytes
4e93adb c625f4c 4e93adb 4cd3056 c625f4c ca8a144 408e821 c625f4c 408e821 4e93adb c625f4c 4e93adb c625f4c 4e93adb c625f4c 4e93adb c625f4c 7dc6d22 c625f4c 7dc6d22 c625f4c 7dc6d22 c625f4c 7dc6d22 c625f4c 1fd7b67 ca8a144 c625f4c ca8a144 c625f4c ca8a144 c625f4c ca8a144 c625f4c ca8a144 c625f4c 4e93adb c625f4c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 |
from rest_framework.views import APIView
from adrf.views import APIView as AsyncAPIView
import tempfile, os
from rest_framework.response import Response
from _utils.resumo_completo_cursor import (
get_llm_summary_answer_by_cursor_complete,
test_ragas,
)
from _utils.resumo_simples_cursor import get_llm_summary_answer_by_cursor
from _utils.utils import DEFAULT_SYSTEM_PROMPT
from .serializer import (
RagasFromTextSerializer,
ResumoCursorCompeltoSerializer,
ResumoPDFSerializer,
ResumoCursorSerializer,
RagasSerializer,
)
from _utils.main import get_llm_answer_summary, get_llm_answer_summary_with_embedding
from setup.environment import default_model
from rest_framework.parsers import MultiPartParser
from drf_spectacular.utils import extend_schema
class ResumoView(APIView):
parser_classes = [MultiPartParser]
@extend_schema(
request=ResumoPDFSerializer,
)
def post(self, request):
serializer = ResumoPDFSerializer(data=request.data)
if serializer.is_valid(raise_exception=True):
listaPDFs = []
data = serializer.validated_data
model = serializer.validated_data.get("model", default_model)
print("serializer.validated_data: ", serializer.validated_data)
for file in serializer.validated_data["files"]:
print("file: ", file)
file.seek(0)
with tempfile.NamedTemporaryFile(
delete=False, suffix=".pdf"
) as temp_file: # Create a temporary file to save the uploaded PDF
for (
chunk
) in (
file.chunks()
): # Write the uploaded file content to the temporary file
temp_file.write(chunk)
temp_file_path = (
temp_file.name
) # Get the path of the temporary file
listaPDFs.append(temp_file_path)
# print('listaPDFs: ', listaPDFs)
system_prompt = data.get("system_prompt", DEFAULT_SYSTEM_PROMPT)
resposta_llm = get_llm_answer_summary(
system_prompt,
data["user_message"],
listaPDFs,
model=model,
isIterativeRefinement=data["iterative_refinement"],
)
for file in listaPDFs:
os.remove(file)
return Response({"resposta": resposta_llm})
class ResumoEmbeddingView(APIView):
parser_classes = [MultiPartParser]
@extend_schema(
request=ResumoPDFSerializer,
)
def post(self, request):
serializer = ResumoPDFSerializer(data=request.data)
if serializer.is_valid(raise_exception=True):
listaPDFs = []
data = serializer.validated_data
model = serializer.validated_data.get("model", default_model)
print("serializer.validated_data: ", serializer.validated_data)
for file in serializer.validated_data["files"]:
file.seek(0)
with tempfile.NamedTemporaryFile(
delete=False, suffix=".pdf"
) as temp_file: # Create a temporary file to save the uploaded PDF
for (
chunk
) in (
file.chunks()
): # Write the uploaded file content to the temporary file
temp_file.write(chunk)
temp_file_path = (
temp_file.name
) # Get the path of the temporary file
listaPDFs.append(temp_file_path)
print("listaPDFs: ", listaPDFs)
system_prompt = data.get("system_prompt", DEFAULT_SYSTEM_PROMPT)
resposta_llm = get_llm_answer_summary_with_embedding(
system_prompt,
data["user_message"],
listaPDFs,
model=model,
isIterativeRefinement=data["iterative_refinement"],
)
for file in listaPDFs:
os.remove(file)
return Response({"resposta": resposta_llm})
class ResumoSimplesCursorView(APIView):
parser_classes = [MultiPartParser]
@extend_schema(
request=ResumoCursorSerializer,
)
def post(self, request):
serializer = ResumoCursorSerializer(data=request.data)
if serializer.is_valid(raise_exception=True):
listaPDFs = []
data = serializer.validated_data
print("\nserializer.validated_data: ", serializer.validated_data)
for file in serializer.validated_data["files"]:
file.seek(0)
with tempfile.NamedTemporaryFile(
delete=False, suffix=".pdf"
) as temp_file: # Create a temporary file to save the uploaded PDF
for (
chunk
) in (
file.chunks()
): # Write the uploaded file content to the temporary file
temp_file.write(chunk)
temp_file_path = (
temp_file.name
) # Get the path of the temporary file
listaPDFs.append(temp_file_path)
print("listaPDFs: ", listaPDFs)
resposta_llm = get_llm_summary_answer_by_cursor(data, listaPDFs)
for file in listaPDFs:
os.remove(file)
return Response({"resposta": resposta_llm})
class ResumoSimplesCursorCompletoView(AsyncAPIView):
parser_classes = [MultiPartParser]
@extend_schema(
request=ResumoCursorCompeltoSerializer,
)
async def post(self, request):
serializer = ResumoCursorCompeltoSerializer(data=request.data)
if serializer.is_valid(raise_exception=True):
print("\n\n\n")
print("serializer.validated_data: ", serializer.validated_data)
print("\n\n\n")
listaPDFs = []
data = serializer.validated_data
print("\nserializer.validated_data: ", serializer.validated_data)
for file in serializer.validated_data["files"]:
file.seek(0)
with tempfile.NamedTemporaryFile(
delete=False, suffix=".pdf"
) as temp_file: # Create a temporary file to save the uploaded PDF
for (
chunk
) in (
file.chunks()
): # Write the uploaded file content to the temporary file
temp_file.write(chunk)
temp_file_path = (
temp_file.name
) # Get the path of the temporary file
listaPDFs.append(temp_file_path)
print("listaPDFs: ", listaPDFs)
# resposta_llm = await get_llm_summary_answer_by_cursor_complete(
# data, listaPDFs
# )
resposta_llm = await get_llm_summary_answer_by_cursor_complete(
data, listaPDFs
)
final = resposta_llm
print("\n\n\n")
print("final: ", final)
for file in listaPDFs:
os.remove(file)
return Response({"resposta": final})
class RagasView(APIView):
parser_classes = [MultiPartParser]
@extend_schema(
request=RagasSerializer,
)
def post(self, request):
serializer = RagasSerializer(data=request.data)
print("\n\n\n")
print("\n\n\n")
print("serializer.data: ", serializer)
listaPDFs = []
if serializer.is_valid(raise_exception=True):
for file in serializer.validated_data["files"]:
file.seek(0)
with tempfile.NamedTemporaryFile(
delete=False, suffix=".pdf"
) as temp_file: # Create a temporary file to save the uploaded PDF
for (
chunk
) in (
file.chunks()
): # Write the uploaded file content to the temporary file
temp_file.write(chunk)
temp_file_path = (
temp_file.name
) # Get the path of the temporary file
listaPDFs.append(temp_file_path)
result = test_ragas(serializer, listaPDFs)
for file in listaPDFs:
os.remove(file)
return Response({"msg": result})
class RagasFromTextView(APIView):
def post(self, request):
serializer = RagasFromTextSerializer(data=request.data)
if serializer.is_valid(raise_exception=True):
from datasets import Dataset
from ragas import evaluate
from ragas.metrics import (
faithfulness,
answer_relevancy,
answer_correctness,
context_precision,
context_recall,
)
import os
from datasets import load_dataset
import pandas as pd
os.environ.get("OPENAI_API_KEY")
df_pandas = pd.read_csv(
"D:/repositorios/projetos-pessoais/projeto-y-backend-hugginf-face-teste-01/vella-backend/_utils/files/ragas_testset.csv"
)
# print(df_pandas["position"]) # Print a specific column
data = {
"user_input": [
"What is the capital of France?",
],
"response": [],
"retrieved_contexts": [],
}
reference = [
"Paris is the capital of France. It is a major European city known for its culture."
]
for x in df_pandas["user_input"]:
data["user_input"].append(x)
for x in df_pandas["reference"]:
reference.append(x)
print("data: ", reference)
for i in range(len(reference)):
serializer.validated_data["user_message"] = data["user_input"][i]
resposta_llm = get_llm_summary_answer_by_cursor_complete(
serializer.validated_data, contexto=reference[i]
)
data["response"].append(resposta_llm["texto_completo"])
lista_reference_contexts = []
for x in resposta_llm["resultado"]:
lista_reference_contexts.append(x["source"]["text"])
data["retrieved_contexts"].append(lista_reference_contexts)
# Convert the data to a Hugging Face Dataset
dataset = Dataset.from_dict(data)
# Define the metrics you want to evaluate
metrics = [
faithfulness,
# answer_relevancy,
# answer_correctness,
# context_precision,
# context_recall,
]
# Evaluate the dataset using the selected metrics
results = evaluate(dataset, metrics)
# results.to_pandas().to_csv("./result.csv")
return Response({"resposta": results.to_pandas().to_string()})
|