Add

app.py

@@ -0,0 +1,50 @@
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+import gradio as gr
+import pickle
+
+torch.autograd.set_grad_enabled(False)
+
+sklearn_model = pickle.load(open('classic_pipeline.pickle', 'rb'))
+
+model_name = "AbstractQbit/electra_large_imdb_htsplice"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+
+def tokenize_with_splicing(text):
+    tokens = tokenizer(text, truncation=False)
+    if len(tokens['input_ids']) > 512:
+        tokens['input_ids'] = tokens['input_ids'][:129] + \
+            [102] + tokens['input_ids'][-382:]
+        tokens['token_type_ids'] = [0]*512
+        tokens['attention_mask'] = [1]*512
+    return tokens
+
+def make_stars(prob):
+    stars = round(1 + prob*9)
+    return '★'*stars + '☆'*(10-stars)
+
+def run_models(review):
+    prob_sklearn = float(sklearn_model.predict_proba([review])[0][1])
+    label_sklearn = 'positive' if prob_sklearn > 0.5 else 'negative'
+    res = f"TF-IDF SVC thinks the review is {label_sklearn} ({100*prob_sklearn:.2f}% positive).\n{make_stars(prob_sklearn):s}\n\n"
+
+    input = tokenize_with_splicing(review).convert_to_tensors('pt', True)
+    output = torch.nn.functional.softmax(model(**input).logits, dim=1)
+    prob_electra = float(output[0][1])
+    label_electra = 'positive' if prob_electra > 0.5 else 'negative'
+    res += f"ELECTRA thinks the review is {label_electra} ({100*prob_electra:.2f}% positive).\n{make_stars(prob_electra):s}"
+
+    return res
+
+
+demo = gr.Interface(
+    fn=run_models, 
+    inputs="text", 
+    outputs="text",
+    title="Movie review classification",
+    allow_flagging='never'
+)
+
+demo.launch()