README.md

metadata

license: apache-2.0
base_model: google/flan-t5-base
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: flan-t5-base-YT-transcript-sum
    results: []

flan-t5-base-YT-transcript-sum

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4118
• Rouge1: 25.4884
• Rouge2: 12.9544
• Rougel: 21.8085
• Rougelsum: 23.7402
• Gen Len: 19.0

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	216	1.5856	23.6948	11.0722	20.1382	22.2371	18.9722
No log	2.0	432	1.4956	23.9099	11.6734	20.4638	22.2083	19.0
1.7743	3.0	648	1.4517	24.8274	12.1689	21.0773	23.1354	19.0
1.7743	4.0	864	1.4334	25.0091	12.4359	21.3687	23.3004	18.9769
1.4377	5.0	1080	1.4175	25.4136	12.792	21.5903	23.5516	18.9815
1.4377	6.0	1296	1.4118	25.4884	12.9544	21.8085	23.7402	19.0
1.2819	7.0	1512	1.4144	25.6073	13.4263	22.1348	24.0203	19.0
1.2819	8.0	1728	1.4181	25.8727	13.5153	22.2189	24.211	19.0
1.2819	9.0	1944	1.4174	25.8817	13.9496	22.4955	24.3655	19.0
1.1716	10.0	2160	1.4193	26.2123	13.7931	22.4856	24.6381	19.0
1.1716	11.0	2376	1.4188	25.7062	13.6106	22.1476	24.1384	19.0
1.0947	12.0	2592	1.4243	26.0097	13.9708	22.4699	24.4448	19.0
1.0947	13.0	2808	1.4307	26.0922	14.1617	22.5681	24.513	18.9954
1.0566	14.0	3024	1.4311	25.9908	13.8775	22.448	24.4268	18.9954
1.0566	15.0	3240	1.4347	26.1238	13.9542	22.5653	24.5333	18.9954

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3