pppp

app.py

@@ -12,27 +12,48 @@ question = "Qual é o maior planeta do sistema solar?"
 
 
 before = datetime.datetime.now()
+
+# Load model directly
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+tokenizer = AutoTokenizer.from_pretrained("01-ai/Yi-1.5-34B-Chat")
+model = AutoModelForCausalLM.from_pretrained("01-ai/Yi-1.5-34B-Chat")
    
-# Use a pipeline as a high-level helper
-from transformers import pipeline
 
-messages = [
-    {"role": "user", "content": question},
-]
+prompt = "Hey, are you conscious? Can you talk to me?"
+inputs = tokenizer(prompt, return_tensors="pt")
 
-print('gerando a saida...')
+# Generate
 
 st.write('gerando a saida...')
+generate_ids = model.generate(inputs.input_ids, max_length=30)
+output = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
 
-pipe = pipeline("text-generation", model="01-ai/Yi-1.5-34B-Chat")
 
-st.write('pipeline...')
+st.write('saída gerada')
 
-output = pipe(messages)
+st.write(output)
 
-st.write('saída gerada...')
+# Use a pipeline as a high-level helper
+# from transformers import pipeline
 
-st.write(output)
+# messages = [
+#     {"role": "user", "content": question},
+# ]
+
+# print('gerando a saida...')
+
+# st.write('gerando a saida...')
+
+# pipe = pipeline("text-generation", model="01-ai/Yi-1.5-34B-Chat")
+
+# st.write('pipeline...')
+
+# output = pipe(messages)
+
+# st.write('saída gerada...')
+
+# st.write(output)