Modelpublic

lobrien001/Experiment1-system1-roberta-base-finetuned-ner

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes7downloads

Model Card

Experiment1-system1-roberta-base-finetuned-ner

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

—Loss: 0.1869
—Precision: 0.9424
—Recall: 0.9335
—F1: 0.9379
—Accuracy: 0.9330

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: 2e-05
—trainbatchsize: 4
—evalbatchsize: 4
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.04	10	0.6735	0.7957	0.8581	0.8257	0.7957
No log	0.07	20	0.5323	0.7957	0.8581	0.8257	0.7957
No log	0.11	30	0.4444	0.8146	0.8785	0.8453	0.8146
No log	0.15	40	0.3747	0.8393	0.8973	0.8674	0.8481
No log	0.19	50	0.3110	0.8734	0.8943	0.8837	0.8777
No log	0.22	60	0.2818	0.8934	0.9031	0.8982	0.8906
No log	0.26	70	0.2628	0.9277	0.8946	0.9108	0.9031
No log	0.3	80	0.2407	0.9190	0.9160	0.9175	0.9133
No log	0.34	90	0.2861	0.9285	0.8775	0.9023	0.8883
No log	0.37	100	0.2523	0.9024	0.9150	0.9086	0.9073
No log	0.41	110	0.2351	0.9195	0.9131	0.9163	0.9122
No log	0.45	120	0.2435	0.9339	0.9060	0.9197	0.9111
No log	0.49	130	0.2365	0.9315	0.9097	0.9205	0.9142
No log	0.52	140	0.2182	0.9345	0.9177	0.9260	0.9202
No log	0.56	150	0.2138	0.9355	0.9182	0.9268	0.9207
No log	0.6	160	0.2140	0.9383	0.9187	0.9284	0.9223
No log	0.63	170	0.2018	0.9397	0.9284	0.9340	0.9285
No log	0.67	180	0.1998	0.9408	0.9284	0.9346	0.9290
No log	0.71	190	0.1930	0.9433	0.9292	0.9362	0.9308
No log	0.75	200	0.1908	0.9420	0.9285	0.9352	0.9300
No log	0.78	210	0.1923	0.9392	0.9275	0.9333	0.9279
No log	0.82	220	0.1891	0.9425	0.9297	0.9361	0.9303
No log	0.86	230	0.1877	0.9449	0.9319	0.9384	0.9326
No log	0.9	240	0.1873	0.9448	0.9319	0.9383	0.9323
No log	0.93	250	0.1868	0.9445	0.9328	0.9386	0.9330
No log	0.97	260	0.1866	0.9429	0.9338	0.9383	0.9333

Framework versions

—Transformers 4.36.2
—Pytorch 1.11.0+cu113
—Datasets 2.19.0
—Tokenizers 0.15.2