Upload app.py

app.py

@@ -1,9 +1,7 @@
 import gradio as gr
-import torch
-import os
-from huggingface_hub import login
-from transformers import AutoModelForSequenceClassification, TrainingArguments, Trainer
-from transformers import AutoTokenizer
+from datasets import load_dataset
+import random
+
 
 README = """
     # Movie Review Score Discriminator
@@ -19,8 +17,6 @@ learning_rate = 5e-5
 batch_size_train = 64
 step = 1900
 
-file_name = "model-{}.pt".format(step)
-state_dict = torch.load(os.path.join(file_name))
 
 id2label = {0: "NEGATIVE", 1: "POSITIVE"}
 label2id = {"NEGATIVE": 0, "POSITIVE": 1}
@@ -28,34 +24,20 @@ label2id = {"NEGATIVE": 0, "POSITIVE": 1}
 
 title = "Movie Review Score Discriminator"
 description = "It is a program that classifies whether it is positive or negative by entering movie reviews. You can choose between the Korean version and the English version."
-examples = ["the greatest musicians ", "cold movie "]
-
 
 
-def tokenized_data(tokenizer, inputs):
-    return tokenizer.batch_encode_plus(
-            inputs,
-            return_tensors="pt",
-            padding="max_length",
-            max_length=64,
-            truncation=True)
+imdb_dataset = load_dataset('imdb')
+examples = []
+# examples = ["the greatest musicians ", "cold movie "]
+for i in range(3):
+    idx = random.randrange(len(imdb_dataset['train']))
+    examples.append(imdb_dataset['train'][idx]['text'])
 
 
-def greet(text):
-    tokenizer = AutoTokenizer.from_pretrained(model_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()
+def fn(text):
+    return "hello, " + text
 
-    with torch.no_grad():
-        # logits.shape = torch.Size([ batch_size, 2 ])
-        logits = model(input_ids=inputs[0], attention_mask=inputs[1]).logits
-
-    return logits
 
 demo1 = gr.Interface.load("models/cardiffnlp/twitter-roberta-base-sentiment", inputs="text", outputs="text", 
                          title=title, theme="peach",
@@ -68,6 +50,15 @@ demo1 = gr.Interface.load("models/cardiffnlp/twitter-roberta-base-sentiment", in
 #                          allow_flagging="auto",
 #                          description=description, examples=examples)
 
+here = gr.Interface(fn,
+                     inputs= gr.inputs.Textbox( lines=1, placeholder=None, default="", label=None),
+                     outputs='text',
+                     title="Sentiment analysis of movie reviews",
+                     description=description,
+                     theme="peach",
+                     allow_flagging="auto",
+                     flagging_dir='flagging records')
+
 
 demo3 = gr.Interface.load("models/mdj1412/movie_review_score_discriminator_eng", inputs="text", outputs="text", 
                          title=title, theme="peach",
@@ -75,4 +66,5 @@ demo3 = gr.Interface.load("models/mdj1412/movie_review_score_discriminator_eng",
                          description=description, examples=examples)
     
 if __name__ == "__main__":
+    # here.launch()
     demo3.launch()