Modelpublic

hawalurahman/idt5-base-qg_adapter_cross

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes2downloads

Model Card

idt5-base-qgadaptercross

This model is a fine-tuned version of muchad/idt5-base on the None dataset. It achieves the following results on the evaluation set:

—Loss: 2.5024
—Rouge1: 0.2095
—Rouge2: 0.0666
—Rougel: 0.1956
—Rougelsum: 0.1956
—Bleu: 0.0302

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

—learning_rate: 0.0001
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bleu
4.2434	1.0	7645	2.8841	0.1168	0.0279	0.1128	0.1128	0.0182
3.9586	2.0	15290	2.6465	0.1805	0.0574	0.1694	0.1694	0.0258
3.8339	3.0	22935	2.5548	0.2063	0.0671	0.1931	0.1931	0.0281
3.7775	4.0	30580	2.5127	0.2073	0.0665	0.1936	0.1937	0.0292
3.7519	5.0	38225	2.5024	0.2095	0.0666	0.1956	0.1956	0.0302

Framework versions

—PEFT 0.13.2
—Transformers 4.46.0
—Pytorch 2.4.0a0+f70bd71a48.nv24.06
—Datasets 3.0.2
—Tokenizers 0.20.1