Spaces:
Sleeping
Sleeping
xxxxxx
commited on
Commit
·
eb36871
1
Parent(s):
dd50be3
update
Browse files
app.py
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
import streamlit as st
|
2 |
-
from transformers import pipeline
|
3 |
import json
|
4 |
from onnxruntime import InferenceSession
|
5 |
-
|
6 |
|
7 |
# 设置页面配置
|
8 |
st.set_page_config(page_title="中文垃圾信息分类器", page_icon="🚫", layout="wide")
|
@@ -13,9 +13,15 @@ def load_classifiers():
|
|
13 |
hf_classifier = pipeline("text-classification", model="app-x/chinese_spam_classifier")
|
14 |
onnx_session = InferenceSession("app-x/chinese_spam_classifier_onnx/model_optimized.onnx")
|
15 |
tokenizer = AutoTokenizer.from_pretrained("app-x/chinese_spam_classifier_onnx")
|
16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
-
hf_classifier, onnx_session, tokenizer = load_classifiers()
|
19 |
|
20 |
st.title("🚫 中文垃圾信息分类器")
|
21 |
st.write("使用两个模型进行中文文本的垃圾信息分类。")
|
@@ -23,6 +29,24 @@ st.write("使用两个模型进行中文文本的垃圾信息分类。")
|
|
23 |
# 创建两列布局
|
24 |
col1, col2 = st.columns([2, 1])
|
25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
with col1:
|
27 |
# 创建文本输入框
|
28 |
text_input = st.text_area("请输入中文文本:", height=200)
|
@@ -30,52 +54,17 @@ with col1:
|
|
30 |
if st.button("分类", key="classify_button"):
|
31 |
if text_input:
|
32 |
with st.spinner("正在分析..."):
|
33 |
-
|
34 |
-
hf_result = hf_classifier(text_input)[0]
|
35 |
-
hf_label = "垃圾信息" if hf_result["label"] == "spam" else "正常信息"
|
36 |
-
hf_confidence = hf_result["score"]
|
37 |
-
|
38 |
-
# ONNX模型分类
|
39 |
-
inputs = tokenizer(text_input, return_tensors="np", padding=True, truncation=True)
|
40 |
-
onnx_result = onnx_session.run(None, dict(inputs))
|
41 |
-
onnx_label = "垃圾信息" if onnx_result[0][0][1] > onnx_result[0][0][0] else "正常信息"
|
42 |
-
onnx_confidence = max(onnx_result[0][0])
|
43 |
-
|
44 |
-
# 创建JSON格式的结果
|
45 |
-
json_result = {
|
46 |
-
"input_text": text_input,
|
47 |
-
"huggingface_model": {
|
48 |
-
"classification": hf_label,
|
49 |
-
"confidence": hf_confidence,
|
50 |
-
"raw_output": hf_result
|
51 |
-
},
|
52 |
-
"onnx_model": {
|
53 |
-
"classification": onnx_label,
|
54 |
-
"confidence": float(onnx_confidence),
|
55 |
-
"raw_output": onnx_result[0].tolist()
|
56 |
-
}
|
57 |
-
}
|
58 |
-
|
59 |
-
# 显示结果
|
60 |
-
st.subheader("HuggingFace模型分类结果:")
|
61 |
-
if hf_label == "垃圾信息":
|
62 |
-
st.error(f"⚠️ {hf_label}")
|
63 |
-
else:
|
64 |
-
st.success(f"✅ {hf_label}")
|
65 |
-
st.write(f"概率: {hf_confidence:.2f}")
|
66 |
-
st.progress(hf_confidence)
|
67 |
-
|
68 |
-
st.subheader("ONNX模型分类结果:")
|
69 |
-
if onnx_label == "垃圾信息":
|
70 |
-
st.error(f"⚠️ {onnx_label}")
|
71 |
-
else:
|
72 |
-
st.success(f"✅ {onnx_label}")
|
73 |
-
st.write(f"概率: {onnx_confidence:.2f}")
|
74 |
-
st.progress(float(onnx_confidence))
|
75 |
|
76 |
-
|
77 |
-
|
78 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
79 |
else:
|
80 |
st.warning("请输入文本后再进行分类。")
|
81 |
|
@@ -104,4 +93,4 @@ with col2:
|
|
104 |
|
105 |
# 添加页脚
|
106 |
st.markdown("---")
|
107 |
-
st.markdown("由 Streamlit 和 Hugging Face 提供支持 | 作者:[app-x]")
|
|
|
1 |
import streamlit as st
|
2 |
+
from transformers import pipeline, AutoTokenizer
|
3 |
import json
|
4 |
from onnxruntime import InferenceSession
|
5 |
+
import numpy as np
|
6 |
|
7 |
# 设置页面配置
|
8 |
st.set_page_config(page_title="中文垃圾信息分类器", page_icon="🚫", layout="wide")
|
|
|
13 |
hf_classifier = pipeline("text-classification", model="app-x/chinese_spam_classifier")
|
14 |
onnx_session = InferenceSession("app-x/chinese_spam_classifier_onnx/model_optimized.onnx")
|
15 |
tokenizer = AutoTokenizer.from_pretrained("app-x/chinese_spam_classifier_onnx")
|
16 |
+
|
17 |
+
# 加载配置文件
|
18 |
+
with open("app-x/chinese_spam_classifier_onnx/config.json", "r") as f:
|
19 |
+
config = json.load(f)
|
20 |
+
|
21 |
+
id2label = config["id2label"]
|
22 |
+
return hf_classifier, onnx_session, tokenizer, id2label
|
23 |
|
24 |
+
hf_classifier, onnx_session, tokenizer, id2label = load_classifiers()
|
25 |
|
26 |
st.title("🚫 中文垃圾信息分类器")
|
27 |
st.write("使用两个模型进行中文文本的垃圾信息分类。")
|
|
|
29 |
# 创建两列布局
|
30 |
col1, col2 = st.columns([2, 1])
|
31 |
|
32 |
+
def classify_text(text):
|
33 |
+
# HuggingFace模型分类
|
34 |
+
hf_result = hf_classifier(text)[0]
|
35 |
+
hf_label = id2label[str(int(hf_result["label"].split("_")[-1]))]
|
36 |
+
hf_confidence = hf_result["score"]
|
37 |
+
|
38 |
+
# ONNX模型分类
|
39 |
+
inputs = tokenizer(text, return_tensors="np", padding=True, truncation=True)
|
40 |
+
onnx_result = onnx_session.run(None, dict(inputs))
|
41 |
+
onnx_probs = onnx_result[0][0]
|
42 |
+
onnx_label = id2label[str(np.argmax(onnx_probs))]
|
43 |
+
onnx_confidence = np.max(onnx_probs)
|
44 |
+
|
45 |
+
return {
|
46 |
+
"hf": {"label": hf_label, "confidence": hf_confidence},
|
47 |
+
"onnx": {"label": onnx_label, "confidence": float(onnx_confidence)}
|
48 |
+
}
|
49 |
+
|
50 |
with col1:
|
51 |
# 创建文本输入框
|
52 |
text_input = st.text_area("请输入中文文本:", height=200)
|
|
|
54 |
if st.button("分类", key="classify_button"):
|
55 |
if text_input:
|
56 |
with st.spinner("正在分析..."):
|
57 |
+
results = classify_text(text_input)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
58 |
|
59 |
+
for model, result in results.items():
|
60 |
+
st.subheader(f"{model.upper()} 模型分类结果:")
|
61 |
+
label = "垃圾信息" if result["label"] == "spam" else "正常信息"
|
62 |
+
if label == "垃圾信息":
|
63 |
+
st.error(f"⚠️ {label}")
|
64 |
+
else:
|
65 |
+
st.success(f"✅ {label}")
|
66 |
+
st.write(f"概率: {result['confidence']:.2f}")
|
67 |
+
st.progress(result['confidence'])
|
68 |
else:
|
69 |
st.warning("请输入文本后再进行分类。")
|
70 |
|
|
|
93 |
|
94 |
# 添加页脚
|
95 |
st.markdown("---")
|
96 |
+
st.markdown("由 Streamlit 和 Hugging Face 提供支持 | 作者:[app-x]")
|