Update app.py

T5 task 3 English-to-German translation

app.py

@@ -222,7 +222,7 @@
 # grad.Interface(generate, inputs=txt, outputs=out).launch()
 
 #-----------------------------------------------------------------------------------
-# 10. Text-to-Text Generation using the T5 model - first use case generates a question given some context.
+# 10. Text-to-Text Generation using the T5 model - Task 1 generates a question given some context.
 
 # A transformer-based architecture that takes a text-to-text approach is referred to as T5, which stands for Text-to-Text Transfer Transformer.
 
@@ -247,35 +247,55 @@
 
 # context=grad.Textbox(lines=10, label="English", placeholder="Context")
 # ans=grad.Textbox(lines=1, label="Answer")
-# out=grad.Textbox(lines=1, label="Genereated Question")
+# out=grad.Textbox(lines=1, label="Generated Question")
 
 # grad.Interface(text2text, inputs=[context,ans], outputs=out).launch()
 
 #-----------------------------------------------------------------------------------
-# 11. Text-to-Text Generation using the T5 model - second use case summarizes a paragraph of text.
+# 11. Text-to-Text Generation using the T5 model - Task 2 summarizes a paragraph of text.
 
-from transformers import AutoTokenizer, AutoModelWithLMHead
+# from transformers import AutoTokenizer, AutoModelWithLMHead
+# import gradio as grad
+
+# text2text_tkn = AutoTokenizer.from_pretrained("deep-learning-analytics/wikihow-t5-small")
+# mdl = AutoModelWithLMHead.from_pretrained("deep-learning-analytics/wikihow-t5-small")
+
+# def text2text_summary(para):
+#     initial_txt = para.strip().replace("\n","")
+#     tkn_text = text2text_tkn.encode(initial_txt, return_tensors="pt")
+#     tkn_ids = mdl.generate(
+#         tkn_text,
+#         max_length=250,
+#         num_beams=5,
+#         repetition_penalty=2.5,
+#         early_stopping=True
+#     )
+#     response = text2text_tkn.decode(tkn_ids[0], skip_special_tokens=True)
+#     return response
+
+# para=grad.Textbox(lines=10, label="Paragraph", placeholder="Copy paragraph")
+# out=grad.Textbox(lines=1, label="Summary")
+
+# grad.Interface(text2text_summary, inputs=para, outputs=out).launch()
+
+#-----------------------------------------------------------------------------------
+# 12. Text-to-Text Generation using the T5 model - Task 3 English-to-German translation.
+
+from transformers import T5ForConditionalGeneration, T5Tokenizer
 import gradio as grad
 
-text2text_tkn = AutoTokenizer.from_pretrained("deep-learning-analytics/wikihow-t5-small")
-mdl = AutoModelWithLMHead.from_pretrained("deep-learning-analytics/wikihow-t5-small")
-
-def text2text_summary(para):
-    initial_txt = para.strip().replace("\n","")
-    tkn_text = text2text_tkn.encode(initial_txt, return_tensors="pt")
-    tkn_ids = mdl.generate(
-        tkn_text,
-        max_length=250,
-        num_beams=5,
-        repetition_penalty=2.5,
-        early_stopping=True
-    )
-    response = text2text_tkn.decode(tkn_ids[0], skip_special_tokens=True)
-    return response
+text2text_tkn= T5Tokenizer.from_pretrained("t5-small")
+mdl = T5ForConditionalGeneration.from_pretrained("t5-small")
 
-para=grad.Textbox(lines=10, label="Paragraph", placeholder="Copy paragraph")
-out=grad.Textbox(lines=1, label="Summary")
+def text2text_translation(text):
+    inp = "translate English to German:: "+text
+    enc = text2text_tkn(inp, return_tensors="pt")
+    tokens = mdl.generate(**enc)
+    response=text2text_tkn.batch_decode(tokens)
+    return response
 
-grad.Interface(text2text_summary, inputs=para, outputs=out).launch()
+para=grad.Textbox(lines=1, label="English Text", placeholder="Text in English")
+out=grad.Textbox(lines=1, label="German Translation")
 
+grad.Interface(text2text_translation, inputs=para, outputs=out).launch()