Edit model card

modelsent_test

This model is a fine-tuned version of albert/albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2310
  • Accuracy: 0.9279
  • F1: 0.9279
  • Precision: 0.9280
  • Recall: 0.9279
  • Accuracy Label Negative: 0.9192
  • Accuracy Label Positive: 0.9361

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
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall Accuracy Label Negative Accuracy Label Positive
0.5427 0.2442 100 0.5228 0.7544 0.7539 0.7602 0.7544 0.8157 0.6970
0.2745 0.4884 200 0.2897 0.8937 0.8937 0.8940 0.8937 0.9028 0.8852
0.2409 0.7326 300 0.3172 0.8992 0.8985 0.9069 0.8992 0.8232 0.9704
0.2717 0.9768 400 0.2341 0.9163 0.9163 0.9169 0.9163 0.9306 0.9030
0.2178 1.2210 500 0.2670 0.9169 0.9169 0.9171 0.9169 0.9230 0.9112
0.2011 1.4652 600 0.2634 0.9145 0.9143 0.9158 0.9145 0.8813 0.9456
0.2179 1.7094 700 0.2657 0.9016 0.9015 0.9027 0.9016 0.8699 0.9314
0.1465 1.9536 800 0.2150 0.9212 0.9210 0.9228 0.9212 0.8851 0.9550
0.1602 2.1978 900 0.2421 0.9261 0.9261 0.9264 0.9261 0.9356 0.9172
0.1293 2.4420 1000 0.2693 0.9181 0.9181 0.9204 0.9181 0.9520 0.8864
0.1023 2.6862 1100 0.2392 0.9236 0.9237 0.9240 0.9236 0.9343 0.9136
0.1663 2.9304 1200 0.2326 0.9267 0.9267 0.9269 0.9267 0.9116 0.9408

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
8
Safetensors
Model size
11.7M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Frenz/modelsent_test

Quantized
(3)
this model