Closed pradeepdev-1995 closed 2 years ago
Hi @pradeepdev-1995 , Thanks for pointing out, I have updated the notebook. The recent version of simpleT5 only supports below methods
# import
from simplet5 import SimpleT5
# instantiate
model = SimpleT5()
# load (supports t5, mt5, byT5 models)
model.from_pretrained("t5","t5-base")
# train
model.train(train_df=train_df, # pandas dataframe with 2 columns: source_text & target_text
eval_df=eval_df, # pandas dataframe with 2 columns: source_text & target_text
source_max_token_len = 512,
target_max_token_len = 128,
batch_size = 8,
max_epochs = 5,
use_gpu = True,
outputdir = "outputs",
early_stopping_patience_epochs = 0,
precision = 32
)
# load trained T5 model
model.load_model("t5","path/to/trained/model/directory", use_gpu=False)
# predict
model.predict("input text for prediction")
@Shivanandroy Thank you Which means convert_and_load_onnx_model is no longer available? If I want to export a model to onnx for faster inference then what should I do?
There are many packages which can help you with quantization and onnx runtime. e.g. fastT5: https://github.com/Ki6an/fastT5
Thank you
Shows the error AttributeError: 'SimpleT5' object has no attribute 'convert_and_load_onnx_model' while running the example notebook provided in the repository
https://github.com/Shivanandroy/simpleT5/blob/main/examples/simpleT5-summarization.ipynb