Hello, I would like to use the T5 network to obtain different outputs with same input, like an intelligent randomness in generation. Example: Run1: Input: input1 Output: output1 Run2: Input: input1 Output: output2 ... How can I do this? Thank you

Hi Gennaro You can use the do_sample parameter to introduce randomness in text generation. Example: [Capture] Once you use sampling, you can even generate multiple responses with one inference request using the num_return_sequences parameter: [Capture2] Hope that helps! Cheers Heiko

T5 randomness in generation

Models

ppujari May 15, 2022, 5:35am 5

Where is the usecase of this?

Topic		Replies	Views
T5 models have non-deterministic outputs even after disabling dropout 🤗Transformers	9	192	September 15, 2024
T5 Model Generate and Model Outputs Vastly Different Beginners	2	835	August 19, 2025
How to generate text with T5Model other than T5ForConditionalGeneration? 🤗Transformers	0	301	September 22, 2022
Text Generation output keep repeat input sentences. Am I missing somethings Beginners	3	948	May 31, 2024
Sample unique sentences via model.generate() 🤗Transformers	0	341	January 29, 2022

T5 randomness in generation

Related topics