natasha model

models/model_n/config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "cointegrated/rubert-tiny-toxicity",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "emb_size": 312,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 312,
+  "id2label": {
+    "0": "non-toxic",
+    "1": "insult",
+    "2": "obscenity",
+    "3": "threat",
+    "4": "dangerous"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 600,
+  "label2id": {
+    "dangerous": 4,
+    "insult": 1,
+    "non-toxic": 0,
+    "obscenity": 2,
+    "threat": 3
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 3,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "multi_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.4",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 29564
+}

models/model_n/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d58739fa66f1975dc1b7e930f6b146283f0593cca1d74a52ddc9643736d0c092
+size 47149380

models/model_n/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/model_n/tokenizer_config.json

@@ -0,0 +1,61 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "mask_token": "[MASK]",
+  "max_length": 512,
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

pages/kdnv_model.py

@@ -7,7 +7,7 @@ import textwrap
 @st.cache_resource()
 def load_model():
     model = GPT2LMHeadModel.from_pretrained('sberbank-ai/rugpt3small_based_on_gpt2')
-    model.load_state_dict(torch.load('models/kdnv_model.pt', map_location=torch.device('cpu')))
+    model.load_state_dict(torch.load('models/kdnv_model.pt', map_location=torch.device('cpu'), weights_only=True))
     return model
 
 

pages/natasha_model.py

@@ -0,0 +1,40 @@
+import streamlit as st
+import torch
+from transformers import BertTokenizer, BertForSequenceClassification
+
+# Путь к директории, где сохранена обученная модель
+model_directory = './models/model_n'
+
+# Загрузка модели и токенизатора из указанной директории
+tokenizer = BertTokenizer.from_pretrained(model_directory)
+model = BertForSequenceClassification.from_pretrained(model_directory)
+
+# # Загрузка модели и токенизатора
+# tokenizer = BertTokenizer.from_pretrained("cointegrated/rubert-tiny-toxicity")
+# model = BertForSequenceClassification.from_pretrained("cointegrated/rubert-tiny-toxicity")
+
+# Перевод модели в режим оценки
+model.eval()
+
+# Функция для оценки токсичности
+def predict_toxicity(text):
+    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits
+        probabilities = torch.nn.functional.softmax(logits, dim=-1)
+    return probabilities[0][1].item()  # Вероятность токсичности
+
+# Streamlit интерфейс
+st.title("Оценка степени токсичности")
+
+# Поле ввода для пользовательского сообщения
+user_input = st.text_area("Введите текст для оценки токсичности:")
+
+# Кнопка для оценки
+if st.button("Оценить токсичность"):
+    if user_input:
+        toxicity_score = predict_toxicity(user_input)
+        st.write(f"Вероятность токсичности: {toxicity_score:.2f}")
+    else:
+        st.write("Пожалуйста, введите текст для оценки.")