codet5-small-ft-v10.1

This model is a fine-tuned version of Salesforce/codet5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2959
• Rouge1: 81.3062
• Rouge2: 75.041
• Rougel: 81.6284
• Rougelsum: 81.0728
• Gen Len: 13.9857

Model description

More information needed

FORMAT

[BUG] reader = ReaderFactory.newXmlReader( null) ; [CONTEXT] outputFile = output.getFile(); } Reader reader; try { if (inputEncoding != null) { if (parser.getType() == Parser.XML_TYPE) { } else { reader = ReaderFactory.newReader(inputFile, inputEncoding); } } else { reader = ReaderFactory.newPlatformReader(inputFile); }

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	19	1.8175	54.4184	41.9095	54.2713	53.6157	12.8857
No log	2.0	38	0.8332	69.024	55.6554	66.6238	66.2558	13.2286
No log	3.0	57	0.5293	64.9854	48.4359	62.4491	61.9973	12.8857
No log	4.0	76	0.4513	70.6432	59.744	69.8409	69.4408	14.3
No log	5.0	95	0.3845	73.7268	64.6645	73.4703	72.9311	14.3571
No log	6.0	114	0.3549	75.2727	66.4769	75.1021	74.6361	14.4571
No log	7.0	133	0.3199	78.5225	72.2594	78.8828	78.3729	13.9
No log	8.0	152	0.3053	79.2229	72.7468	79.6079	79.1507	14.2429
No log	9.0	171	0.2998	80.4535	73.8457	80.587	80.1368	14.1571
No log	10.0	190	0.2959	81.3062	75.041	81.6284	81.0728	13.9857

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1