MCQ-QuestionGen

This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3630
• Rouge1: 0.5451
• Rouge2: 0.3214
• Rougel: 0.5019
• Bleu: 0.2368
• Em: 0.0504
• F1: 0.5218

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
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Bleu	Em	F1
2.8703	1.0	2937	1.2938	0.5324	0.3045	0.489	0.2175	0.0387	0.5075
2.6182	2.0	5874	1.2576	0.5324	0.3067	0.49	0.2197	0.0433	0.5078
2.3196	3.0	8811	1.2556	0.5382	0.3136	0.4959	0.228	0.0431	0.5144
1.9782	4.0	11748	1.2703	0.5461	0.3219	0.5033	0.2368	0.0467	0.5226
1.9494	5.0	14685	1.2857	0.5456	0.3214	0.5036	0.2349	0.0477	0.5223
1.7425	6.0	17622	1.3035	0.5455	0.322	0.5025	0.2369	0.05	0.5222
1.6953	7.0	20559	1.3217	0.5461	0.322	0.5029	0.2365	0.0483	0.5226
1.5681	8.0	23496	1.3481	0.5458	0.3218	0.503	0.2381	0.05	0.5222
1.4350	9.0	26433	1.3630	0.5451	0.3214	0.5019	0.2368	0.0504	0.5218

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2