Upload 5 files

app.py

@@ -28,7 +28,7 @@ description = "It is a program that classifies whether it is positive or negativ
 
 def tokenized_data(tokenizer, inputs):
     return tokenizer.batch_encode_plus(
-        inputs,
+        [inputs],
         return_tensors="pt",
         padding="max_length",
         max_length=64,
@@ -44,35 +44,42 @@ examples = []
 df = pd.read_csv('examples.csv', sep='\t', index_col='Unnamed: 0')
 for i in range(2):
     idx = random.randint(0, 50)
-    examples.append(df.iloc[idx, 0])
-    examples.append(df.iloc[idx, 1])
-
-
-
-
-model_kor = gr.Interface.load("models/cardiffnlp/twitter-roberta-base-sentiment")
-model_eng = gr.Interface.load("models/mdj1412/movie_review_score_discriminator_eng")
-
-
-
-def builder(version, inputs):
-    if version == 'Eng':
-        model_name = "roberta-base"
-        step = 1900
-
+    examples.append(['Eng', df.iloc[idx, 0]])
+    examples.append(['Kor', df.iloc[idx, 1]])
+
+
+eng_model_name = "roberta-base"
+eng_step = 1900
+eng_tokenizer = AutoTokenizer.from_pretrained(eng_model_name)
+eng_file_name = "{}-{}.pt".format(eng_model_name, eng_step)
+eng_state_dict = torch.load(eng_file_name)
+eng_model = AutoModelForSequenceClassification.from_pretrained(
+    eng_model_name, num_labels=2, id2label=id2label, label2id=label2id,
+    state_dict=eng_state_dict
+)
+
+
+kor_model_name = "klue_roberta-small"
+kor_step = 2400
+kor_tokenizer = AutoTokenizer.from_pretrained(kor_model_name.replace('_', '/'))
+kor_file_name = "{}-{}.pt".format(kor_model_name, kor_step)
+kor_state_dict = torch.load(kor_file_name)
+kor_model = AutoModelForSequenceClassification.from_pretrained(
+    kor_model_name.replace('_', '/'), num_labels=2, id2label=id2label, label2id=label2id,
+    state_dict=kor_state_dict
+)
+
+
+def builder(lang, text):
+    if lang == 'Eng':
+        model = eng_model
+        tokenizer = eng_tokenizer
     else:
-        model_name = "klue/roberta-small"
-        step = 2400
+        model = kor_model
+        tokenizer = kor_tokenizer
         
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
-    inputs = tokenized_data(tokenizer, inputs)
-    file_name = "{}-{}.pt".format(model_name, step)
-    state_dict = torch.load(file_name)
-    model = AutoModelForSequenceClassification.from_pretrained(
-        model_name, num_labels=2, id2label=id2label, label2id=label2id,
-        state_dict=state_dict
-    )
-
+    inputs = tokenized_data(tokenizer, text)
+    
     model.eval()
     with torch.no_grad():
         logits = model(input_ids=inputs['input_ids'], 
@@ -84,11 +91,11 @@ def builder(version, inputs):
 
 
 def builder2(inputs):
-    return model_eng(inputs)
+    return eng_model(inputs)
 
 
 demo = gr.Interface(builder, inputs=[gr.inputs.Dropdown(['Eng', 'Kor']), "text"], outputs="text", 
-                            title=title, description=description, examples=[examples])
+                            title=title, description=description, examples=examples)
 
 # demo2 = gr.Interface(builder2, inputs="text", outputs="text", 
 #                          title=title, theme="peach",
@@ -101,5 +108,6 @@ demo = gr.Interface(builder, inputs=[gr.inputs.Dropdown(['Eng', 'Kor']), "text"]
 #                          description=description, examples=examples)
     
 if __name__ == "__main__":
+    # print(examples)
     demo.launch()
     # demo3.launch()

klue_roberta-small-2400.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b572a576888999c3696750507168b1ec8c194b93e3b0a5fb69d5932cb61a410
+size 272408049