Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -26,27 +26,55 @@ try:
|
|
| 26 |
except Exception as e:
|
| 27 |
print(f"❌ Erro ao inicializar Firebase: {e}")
|
| 28 |
|
| 29 |
-
# ======
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
try:
|
| 31 |
audio_pipeline = pipeline(
|
| 32 |
task="audio-classification",
|
| 33 |
-
model=
|
| 34 |
)
|
| 35 |
-
print("✅ audio_pipeline carregado
|
| 36 |
except Exception as e:
|
| 37 |
-
print(f"❌ Erro ao carregar audio_pipeline: {e}")
|
| 38 |
audio_pipeline = None
|
| 39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
# ====== MAPEAMENTO DE EMOÇÕES (ING->PT) ======
|
|
|
|
| 41 |
emotion_labels = {
|
| 42 |
"angry": "raiva",
|
|
|
|
| 43 |
"disgust": "insegurança",
|
| 44 |
"fearful": "ansiedade",
|
| 45 |
"fear": "ansiedade",
|
| 46 |
"happy": "alegria",
|
|
|
|
| 47 |
"neutral": "neutro",
|
| 48 |
"sad": "tristeza",
|
|
|
|
| 49 |
"surprised": "surpreso",
|
|
|
|
| 50 |
}
|
| 51 |
|
| 52 |
# ====== SUGESTÕES ======
|
|
@@ -63,7 +91,7 @@ def gerar_sugestao(emotion_pt):
|
|
| 63 |
}
|
| 64 |
return sugestoes.get(emotion_pt, "Mantenha o equilíbrio emocional e cuide de você mesmo.")
|
| 65 |
|
| 66 |
-
# ====== FALLBACK
|
| 67 |
EMOTION_KEYWORDS = {
|
| 68 |
"tristeza": ["triste","desanimado","melancólico","chateado","solitário","deprimido","abatido","infeliz","desmotivado"],
|
| 69 |
"ansiedade": ["ansioso","preocupado","nervoso","tenso","inquieto","aflito","alarmado","sobrecarregado","inseguro","apreensivo"],
|
|
@@ -89,52 +117,7 @@ def fallback_emotion(text):
|
|
| 89 |
"debug": "Fallback ativado"
|
| 90 |
}
|
| 91 |
|
| 92 |
-
# ======
|
| 93 |
-
def hybrid_emotion(text, result):
|
| 94 |
-
text_lower = (text or "").lower()
|
| 95 |
-
detected = result.get("emotion", "neutro")
|
| 96 |
-
max_matches = 0
|
| 97 |
-
|
| 98 |
-
for emo, keywords in EMOTION_KEYWORDS.items():
|
| 99 |
-
matches = sum(2 for w in keywords if w in text_lower)
|
| 100 |
-
if matches > max_matches:
|
| 101 |
-
max_matches = matches
|
| 102 |
-
if emo != detected:
|
| 103 |
-
detected = emo
|
| 104 |
-
|
| 105 |
-
confidence = result.get("confidence", 0.0)
|
| 106 |
-
if detected != result.get("emotion"):
|
| 107 |
-
confidence = 0.7 + max_matches * 0.05
|
| 108 |
-
confidence = min(confidence, 1.0)
|
| 109 |
-
|
| 110 |
-
return {
|
| 111 |
-
"status": "ok",
|
| 112 |
-
"emotion": detected,
|
| 113 |
-
"emode": [detected],
|
| 114 |
-
"confidence": round(confidence, 2),
|
| 115 |
-
"probabilities": result.get("probabilities", {detected: 1.0}),
|
| 116 |
-
"suggestion": result.get("suggestion", gerar_sugestao(detected)),
|
| 117 |
-
"debug": result.get("debug", "Híbrido aplicado")
|
| 118 |
-
}
|
| 119 |
-
|
| 120 |
-
# ====== HELPERS PARA ÁUDIO ======
|
| 121 |
-
def save_bytes_to_tempfile(bbytes, suffix=".wav"):
|
| 122 |
-
fd, path = tempfile.mkstemp(suffix=suffix)
|
| 123 |
-
os.close(fd)
|
| 124 |
-
with open(path, "wb") as f:
|
| 125 |
-
f.write(bbytes)
|
| 126 |
-
return path
|
| 127 |
-
|
| 128 |
-
def fetch_url_to_tempfile(url):
|
| 129 |
-
r = requests.get(url, timeout=15)
|
| 130 |
-
r.raise_for_status()
|
| 131 |
-
content_type = r.headers.get("content-type", "")
|
| 132 |
-
suffix = ".wav"
|
| 133 |
-
if "mpeg" in content_type or "mp3" in content_type:
|
| 134 |
-
suffix = ".mp3"
|
| 135 |
-
return save_bytes_to_tempfile(r.content, suffix=suffix)
|
| 136 |
-
|
| 137 |
-
# ====== UTIL: Softmax com temperatura para calibrar probabilidades ======
|
| 138 |
def tempered_softmax(scores_dict, temperature=1.0):
|
| 139 |
labels = list(scores_dict.keys())
|
| 140 |
vals = np.array([scores_dict[l] for l in labels], dtype=float)
|
|
@@ -145,7 +128,6 @@ def tempered_softmax(scores_dict, temperature=1.0):
|
|
| 145 |
probs = exps / np.sum(exps)
|
| 146 |
return dict(zip(labels, probs))
|
| 147 |
|
| 148 |
-
# ====== UTIL: média/união de probabilidades ======
|
| 149 |
def average_probabilities(list_of_prob_dicts):
|
| 150 |
all_keys = set()
|
| 151 |
for d in list_of_prob_dicts:
|
|
@@ -158,13 +140,30 @@ def average_probabilities(list_of_prob_dicts):
|
|
| 158 |
avg[k] += d.get(k, 0.0)
|
| 159 |
n = len(list_of_prob_dicts)
|
| 160 |
for k in avg:
|
| 161 |
-
avg[k] =
|
| 162 |
total = sum(avg.values()) or 1.0
|
| 163 |
for k in avg:
|
| 164 |
-
avg[k] =
|
| 165 |
return avg
|
| 166 |
|
| 167 |
-
# ======
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 168 |
@app.route("/analyze", methods=["POST"])
|
| 169 |
def analyze():
|
| 170 |
try:
|
|
@@ -172,7 +171,7 @@ def analyze():
|
|
| 172 |
audio_bytes = None
|
| 173 |
data = None
|
| 174 |
|
| 175 |
-
#
|
| 176 |
if "file" in request.files:
|
| 177 |
f = request.files["file"]
|
| 178 |
audio_bytes = f.read()
|
|
@@ -181,14 +180,12 @@ def analyze():
|
|
| 181 |
data = request.get_json(silent=True)
|
| 182 |
except Exception:
|
| 183 |
data = None
|
| 184 |
-
|
| 185 |
if data:
|
| 186 |
if "audio_base64" in data:
|
| 187 |
audio_bytes = base64.b64decode(data["audio_base64"])
|
| 188 |
elif "audio_url" in data:
|
| 189 |
audio_path = fetch_url_to_tempfile(data["audio_url"])
|
| 190 |
elif "text" in data and (not audio_bytes and not audio_path):
|
| 191 |
-
# apenas texto -> fallback textual
|
| 192 |
return jsonify(fallback_emotion(data["text"]))
|
| 193 |
|
| 194 |
if audio_bytes:
|
|
@@ -202,60 +199,50 @@ def analyze():
|
|
| 202 |
return jsonify(fallback_emotion(data["text"]))
|
| 203 |
return jsonify({"error": "Modelo de áudio indisponível no momento."}), 500
|
| 204 |
|
| 205 |
-
# ----------
|
| 206 |
-
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
run_probs = []
|
| 211 |
-
raw_runs = []
|
| 212 |
|
| 213 |
for topk in topk_list:
|
| 214 |
try:
|
| 215 |
raw_result = audio_pipeline(audio_path, top_k=topk)
|
| 216 |
-
# normalizar formato: raw_result é lista de dicts [{'label':..., 'score':...}, ...]
|
| 217 |
probs = {}
|
|
|
|
| 218 |
for item in raw_result:
|
| 219 |
-
|
| 220 |
-
if
|
| 221 |
-
|
| 222 |
-
probs[
|
| 223 |
if probs:
|
| 224 |
run_probs.append(probs)
|
| 225 |
raw_runs.append({"top_k": topk, "raw": raw_result})
|
| 226 |
except Exception as e:
|
| 227 |
-
|
| 228 |
-
print(f"⚠️ audio_pipeline falhou no top_k={topk}: {e}")
|
| 229 |
|
| 230 |
if not run_probs:
|
| 231 |
return jsonify({"error": "Modelo não retornou rótulos em nenhuma tentativa."}), 500
|
| 232 |
|
| 233 |
-
# 1) média das probabilidades (por rótulo) entre as execuções
|
| 234 |
avg_probs = average_probabilities(run_probs)
|
| 235 |
|
| 236 |
-
#
|
| 237 |
-
|
| 238 |
-
calibrated_probs = tempered_softmax(avg_probs, temperature=temp)
|
| 239 |
|
| 240 |
-
#
|
| 241 |
-
|
| 242 |
-
min_prob_threshold = float(os.getenv("MIN_LABEL_PROB", 0.02)) # 2% por padrão
|
| 243 |
-
filtered = {k: v if v >= min_prob_threshold else 0.0 for k, v in calibrated_probs.items()}
|
| 244 |
totalf = sum(filtered.values()) or 1.0
|
| 245 |
normalized = {k: (v / totalf) for k, v in filtered.items()}
|
| 246 |
|
| 247 |
-
# escolher rótulo final
|
| 248 |
top_label = max(normalized, key=normalized.get)
|
| 249 |
top_score = normalized[top_label]
|
| 250 |
|
| 251 |
-
# map to portuguese
|
| 252 |
emotion_pt = emotion_labels.get(top_label, "desconhecido")
|
| 253 |
-
|
| 254 |
-
# regra de negócio: tristeza muito forte -> depressão
|
| 255 |
-
if emotion_pt == "tristeza" and top_score >= float(os.getenv("DEPRESSION_THRESHOLD", 0.92)):
|
| 256 |
emotion_pt = "depressão"
|
| 257 |
|
| 258 |
-
# montar probabilidades para output (mapeadas p/ pt)
|
| 259 |
probabilities_pt = { emotion_labels.get(k, k): round(float(v), 3) for k, v in normalized.items() }
|
| 260 |
|
| 261 |
base_result = {
|
|
@@ -266,21 +253,15 @@ def analyze():
|
|
| 266 |
"probabilities": probabilities_pt,
|
| 267 |
"suggestion": gerar_sugestao(emotion_pt),
|
| 268 |
"debug": {
|
|
|
|
| 269 |
"runs": raw_runs,
|
| 270 |
"avg_probs": {k: round(float(v), 4) for k, v in avg_probs.items()},
|
| 271 |
-
"
|
| 272 |
-
"
|
| 273 |
}
|
| 274 |
}
|
| 275 |
|
| 276 |
-
|
| 277 |
-
text_for_hybrid = None
|
| 278 |
-
if data and "text" in data:
|
| 279 |
-
text_for_hybrid = data["text"]
|
| 280 |
-
|
| 281 |
-
final_result = hybrid_emotion(text_for_hybrid, base_result) if text_for_hybrid else base_result
|
| 282 |
-
|
| 283 |
-
return jsonify(final_result)
|
| 284 |
|
| 285 |
except Exception as e:
|
| 286 |
print(f"❌ Erro na rota /analyze: {e}")
|
|
@@ -293,4 +274,4 @@ def analyze():
|
|
| 293 |
pass
|
| 294 |
|
| 295 |
if __name__ == "__main__":
|
| 296 |
-
app.run(host="0.0.0.0", port=int(os.getenv("PORT", 7860)))
|
|
|
|
| 26 |
except Exception as e:
|
| 27 |
print(f"❌ Erro ao inicializar Firebase: {e}")
|
| 28 |
|
| 29 |
+
# ====== CONFIGS AJUSTÁVEIS (env vars) ======
|
| 30 |
+
# Modelo recomendado para PT (substitua se quiser um checkpoint em inglês)
|
| 31 |
+
AUDIO_SER_MODEL = os.getenv("AUDIO_SER_MODEL", "alefiury/wav2vec2-xls-r-300m-pt-br-spontaneous-speech-emotion-recognition")
|
| 32 |
+
# Ex.: set ENABLE_ASR=true para ativar ASR (pode consumir muita memória)
|
| 33 |
+
ENABLE_ASR = os.getenv("ENABLE_ASR", "false").lower() in ("1", "true", "yes")
|
| 34 |
+
AUDIO_TOPK_RUNS = os.getenv("AUDIO_TOPK_RUNS", "10,15,20") # exemplo: "10,15,20"
|
| 35 |
+
AUDIO_SOFTMAX_TEMP = float(os.getenv("AUDIO_SOFTMAX_TEMP", "0.6"))
|
| 36 |
+
MIN_LABEL_PROB = float(os.getenv("MIN_LABEL_PROB", "0.02"))
|
| 37 |
+
DEPRESSION_THRESHOLD = float(os.getenv("DEPRESSION_THRESHOLD", "0.92"))
|
| 38 |
+
|
| 39 |
+
# ====== PIPELINE: modelo SER (wav2vec2 finetuned) ======
|
| 40 |
try:
|
| 41 |
audio_pipeline = pipeline(
|
| 42 |
task="audio-classification",
|
| 43 |
+
model=AUDIO_SER_MODEL
|
| 44 |
)
|
| 45 |
+
print(f"✅ audio_pipeline carregado: {AUDIO_SER_MODEL}")
|
| 46 |
except Exception as e:
|
| 47 |
+
print(f"❌ Erro ao carregar audio_pipeline ({AUDIO_SER_MODEL}): {e}")
|
| 48 |
audio_pipeline = None
|
| 49 |
|
| 50 |
+
# Opcional: ASR (desativado por padrão para economia de recursos)
|
| 51 |
+
asr_pipeline = None
|
| 52 |
+
if ENABLE_ASR:
|
| 53 |
+
try:
|
| 54 |
+
asr_pipeline = pipeline(
|
| 55 |
+
task="automatic-speech-recognition",
|
| 56 |
+
model="openai/whisper-large-v3"
|
| 57 |
+
)
|
| 58 |
+
print("✅ asr_pipeline carregado (ENABLE_ASR=true).")
|
| 59 |
+
except Exception as e:
|
| 60 |
+
print(f"⚠️ ASR indisponível: {e}")
|
| 61 |
+
asr_pipeline = None
|
| 62 |
+
|
| 63 |
# ====== MAPEAMENTO DE EMOÇÕES (ING->PT) ======
|
| 64 |
+
# OBS: cada modelo pode usar rótulos ligeiramente diferentes; padronizamos para estes
|
| 65 |
emotion_labels = {
|
| 66 |
"angry": "raiva",
|
| 67 |
+
"anger": "raiva",
|
| 68 |
"disgust": "insegurança",
|
| 69 |
"fearful": "ansiedade",
|
| 70 |
"fear": "ansiedade",
|
| 71 |
"happy": "alegria",
|
| 72 |
+
"joy": "alegria",
|
| 73 |
"neutral": "neutro",
|
| 74 |
"sad": "tristeza",
|
| 75 |
+
"sadness": "tristeza",
|
| 76 |
"surprised": "surpreso",
|
| 77 |
+
"surprise": "surpreso",
|
| 78 |
}
|
| 79 |
|
| 80 |
# ====== SUGESTÕES ======
|
|
|
|
| 91 |
}
|
| 92 |
return sugestoes.get(emotion_pt, "Mantenha o equilíbrio emocional e cuide de você mesmo.")
|
| 93 |
|
| 94 |
+
# ====== FALLBACK POR TEXTO ======
|
| 95 |
EMOTION_KEYWORDS = {
|
| 96 |
"tristeza": ["triste","desanimado","melancólico","chateado","solitário","deprimido","abatido","infeliz","desmotivado"],
|
| 97 |
"ansiedade": ["ansioso","preocupado","nervoso","tenso","inquieto","aflito","alarmado","sobrecarregado","inseguro","apreensivo"],
|
|
|
|
| 117 |
"debug": "Fallback ativado"
|
| 118 |
}
|
| 119 |
|
| 120 |
+
# ====== UTIL: softmax temperado ======
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
def tempered_softmax(scores_dict, temperature=1.0):
|
| 122 |
labels = list(scores_dict.keys())
|
| 123 |
vals = np.array([scores_dict[l] for l in labels], dtype=float)
|
|
|
|
| 128 |
probs = exps / np.sum(exps)
|
| 129 |
return dict(zip(labels, probs))
|
| 130 |
|
|
|
|
| 131 |
def average_probabilities(list_of_prob_dicts):
|
| 132 |
all_keys = set()
|
| 133 |
for d in list_of_prob_dicts:
|
|
|
|
| 140 |
avg[k] += d.get(k, 0.0)
|
| 141 |
n = len(list_of_prob_dicts)
|
| 142 |
for k in avg:
|
| 143 |
+
avg[k] /= n
|
| 144 |
total = sum(avg.values()) or 1.0
|
| 145 |
for k in avg:
|
| 146 |
+
avg[k] /= total
|
| 147 |
return avg
|
| 148 |
|
| 149 |
+
# ====== HELPERS ÁUDIO ======
|
| 150 |
+
def save_bytes_to_tempfile(bbytes, suffix=".wav"):
|
| 151 |
+
fd, path = tempfile.mkstemp(suffix=suffix)
|
| 152 |
+
os.close(fd)
|
| 153 |
+
with open(path, "wb") as f:
|
| 154 |
+
f.write(bbytes)
|
| 155 |
+
return path
|
| 156 |
+
|
| 157 |
+
def fetch_url_to_tempfile(url):
|
| 158 |
+
r = requests.get(url, timeout=15)
|
| 159 |
+
r.raise_for_status()
|
| 160 |
+
content_type = r.headers.get("content-type", "")
|
| 161 |
+
suffix = ".wav"
|
| 162 |
+
if "mpeg" in content_type or "mp3" in content_type:
|
| 163 |
+
suffix = ".mp3"
|
| 164 |
+
return save_bytes_to_tempfile(r.content, suffix=suffix)
|
| 165 |
+
|
| 166 |
+
# ====== ROTA /analyze ======
|
| 167 |
@app.route("/analyze", methods=["POST"])
|
| 168 |
def analyze():
|
| 169 |
try:
|
|
|
|
| 171 |
audio_bytes = None
|
| 172 |
data = None
|
| 173 |
|
| 174 |
+
# receber multipart/file ou json
|
| 175 |
if "file" in request.files:
|
| 176 |
f = request.files["file"]
|
| 177 |
audio_bytes = f.read()
|
|
|
|
| 180 |
data = request.get_json(silent=True)
|
| 181 |
except Exception:
|
| 182 |
data = None
|
|
|
|
| 183 |
if data:
|
| 184 |
if "audio_base64" in data:
|
| 185 |
audio_bytes = base64.b64decode(data["audio_base64"])
|
| 186 |
elif "audio_url" in data:
|
| 187 |
audio_path = fetch_url_to_tempfile(data["audio_url"])
|
| 188 |
elif "text" in data and (not audio_bytes and not audio_path):
|
|
|
|
| 189 |
return jsonify(fallback_emotion(data["text"]))
|
| 190 |
|
| 191 |
if audio_bytes:
|
|
|
|
| 199 |
return jsonify(fallback_emotion(data["text"]))
|
| 200 |
return jsonify({"error": "Modelo de áudio indisponível no momento."}), 500
|
| 201 |
|
| 202 |
+
# ----- Ensemble interno: múltiplas runs com diferentes top_k -----
|
| 203 |
+
topk_list = [int(x) for x in AUDIO_TOPK_RUNS.split(",") if x.strip().isdigit()]
|
| 204 |
+
if not topk_list:
|
| 205 |
+
topk_list = [10, 15, 20]
|
| 206 |
+
|
| 207 |
+
run_probs = []
|
| 208 |
+
raw_runs = []
|
| 209 |
|
| 210 |
for topk in topk_list:
|
| 211 |
try:
|
| 212 |
raw_result = audio_pipeline(audio_path, top_k=topk)
|
|
|
|
| 213 |
probs = {}
|
| 214 |
+
# raw_result é lista de dicts
|
| 215 |
for item in raw_result:
|
| 216 |
+
lbl = item.get("label", "").lower()
|
| 217 |
+
if lbl == "fear":
|
| 218 |
+
lbl = "fearful"
|
| 219 |
+
probs[lbl] = float(item.get("score", 0.0))
|
| 220 |
if probs:
|
| 221 |
run_probs.append(probs)
|
| 222 |
raw_runs.append({"top_k": topk, "raw": raw_result})
|
| 223 |
except Exception as e:
|
| 224 |
+
print(f"⚠️ audio_pipeline falhou top_k={topk}: {e}")
|
|
|
|
| 225 |
|
| 226 |
if not run_probs:
|
| 227 |
return jsonify({"error": "Modelo não retornou rótulos em nenhuma tentativa."}), 500
|
| 228 |
|
|
|
|
| 229 |
avg_probs = average_probabilities(run_probs)
|
| 230 |
|
| 231 |
+
# recalibrar com temperatura (mais baixa => mais confiante)
|
| 232 |
+
calibrated = tempered_softmax(avg_probs, temperature=AUDIO_SOFTMAX_TEMP)
|
|
|
|
| 233 |
|
| 234 |
+
# filtrar rótulos fracos
|
| 235 |
+
filtered = {k: (v if v >= MIN_LABEL_PROB else 0.0) for k, v in calibrated.items()}
|
|
|
|
|
|
|
| 236 |
totalf = sum(filtered.values()) or 1.0
|
| 237 |
normalized = {k: (v / totalf) for k, v in filtered.items()}
|
| 238 |
|
|
|
|
| 239 |
top_label = max(normalized, key=normalized.get)
|
| 240 |
top_score = normalized[top_label]
|
| 241 |
|
|
|
|
| 242 |
emotion_pt = emotion_labels.get(top_label, "desconhecido")
|
| 243 |
+
if emotion_pt == "tristeza" and top_score >= DEPRESSION_THRESHOLD:
|
|
|
|
|
|
|
| 244 |
emotion_pt = "depressão"
|
| 245 |
|
|
|
|
| 246 |
probabilities_pt = { emotion_labels.get(k, k): round(float(v), 3) for k, v in normalized.items() }
|
| 247 |
|
| 248 |
base_result = {
|
|
|
|
| 253 |
"probabilities": probabilities_pt,
|
| 254 |
"suggestion": gerar_sugestao(emotion_pt),
|
| 255 |
"debug": {
|
| 256 |
+
"model": AUDIO_SER_MODEL,
|
| 257 |
"runs": raw_runs,
|
| 258 |
"avg_probs": {k: round(float(v), 4) for k, v in avg_probs.items()},
|
| 259 |
+
"calibrated": {k: round(float(v), 4) for k, v in calibrated.items()},
|
| 260 |
+
"normalized": {k: round(float(v), 4) for k, v in normalized.items()}
|
| 261 |
}
|
| 262 |
}
|
| 263 |
|
| 264 |
+
return jsonify(base_result)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
|
| 266 |
except Exception as e:
|
| 267 |
print(f"❌ Erro na rota /analyze: {e}")
|
|
|
|
| 274 |
pass
|
| 275 |
|
| 276 |
if __name__ == "__main__":
|
| 277 |
+
app.run(host="0.0.0.0", port=int(os.getenv("PORT", 7860)))
|