Upload 3 files

app.py

@@ -0,0 +1,33 @@
+import gradio as gr
+import torch
+from transformers import DistilBertTokenizerFast, DistilBertForSequenceClassification, AutoConfig
+import os
+
+model_name = "distilbert-base-uncased"
+model_path = "cefr_model.pth"
+num_labels = 6
+level_mapping = {0: 'A1', 1: 'A2', 2: 'B1', 3: 'B2', 4: 'C1', 5: 'C2'}
+
+tokenizer = DistilBertTokenizerFast.from_pretrained(model_name)
+config = AutoConfig.from_pretrained(model_name, num_labels=num_labels)
+model = DistilBertForSequenceClassification.from_pretrained(model_name, config=config)
+model.load_state_dict(torch.load(model_path, map_location=torch.device('cpu')))
+model.eval() # Set the model to evaluation mode
+
+def predict_cefr(text):
+    inputs = tokenizer(text, padding=True, truncation=True, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model(**inputs)
+        probabilities = torch.nn.functional.softmax(outputs.logits, dim=-1)
+        predicted_class = torch.argmax(probabilities).item()
+    return level_mapping[predicted_class]
+
+iface = gr.Interface(
+    fn=predict_cefr,
+    inputs=gr.Textbox(lines=2, placeholder="Enter text here..."),
+    outputs="text",
+    title="CEFR Level Prediction",
+    description="Enter some English text and I'll predict its CEFR level!"
+)
+
+iface.launch()

cefr_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fa983346b4c4c868a6393c3487ef1db0a480e7b7b2d8bf8c06ad29a4dc91770
+size 267895464

requirements.txt

@@ -0,0 +1,7 @@
+transformers
+torch
+pandas
+scikit-learn
+gradio
+datasets
+huggingface_hub