Init

.gitignore

@@ -0,0 +1,3 @@
+.venv/
+pyproject.toml
+flagged/

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+from transformers import BertTokenizerFast, BertForSequenceClassification
+import torch
+
+model = BertForSequenceClassification.from_pretrained('./ch-sent-check-model')
+tokenizer = BertTokenizerFast.from_pretrained('./ch-sent-check-model')
+
+def judge(sentence):
+    input_ids = tokenizer(sentence,return_tensors='pt')['input_ids']
+    out = model(input_ids)
+    logits = out.logits
+    pred = torch.argmax(logits,dim=-1).item()
+    pred_text = 'Incorrect' if pred == 0 else 'Correct'
+    return pred_text
+
+iface = gr.Interface(
+    fn=judge, 
+    inputs=gr.Textbox(
+        label="Initial text",
+        lines=1,
+        value="請注意用字的鄭確性",
+    ), 
+    outputs="text"
+)
+iface.launch()

ch-sent-check-model/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "bert-base-chinese",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 21128
+}

ch-sent-check-model/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8b7cb387f207d502b34d79ae10d12e7e7223c0078a90300002400044bc437fb
+size 409144941

ch-sent-check-model/special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

ch-sent-check-model/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "bert-base-chinese", "tokenizer_class": "BertTokenizer"}

requirements.txt

@@ -0,0 +1,3 @@
+transformers==4.*.*
+torch==1.12.*
+gradio