Model save

README.md

@@ -1,37 +1,72 @@
 ---
-language: en
+library_name: transformers
 tags:
-- bert
-- classification
-- pytorch
-pipeline_tag: text-classification
+- generated_from_trainer
+metrics:
+- accuracy
+- f1
+- precision
+- recall
+model-index:
+- name: bert-clf-biencoder-focal_loss
+  results: []
 ---
 
-# BiEncoder Classification Model
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
 
-This model is a BiEncoder architecture based on BERT for text pair classification.
+# bert-clf-biencoder-focal_loss
 
-## Model Details
-- Base Model: bert-base-uncased
-- Architecture: BiEncoder with BERT base
-- Number of classes: 4
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1136
+- Accuracy: 0.6602
+- F1: 0.6596
+- Precision: 0.6642
+- Recall: 0.6602
 
-## Usage
+## Model description
 
-```python
-from transformers import AutoTokenizer
-import torch
+More information needed
 
-# Load tokenizer
-tokenizer = AutoTokenizer.from_pretrained("minoosh/bert-clf-biencoder-focal_loss")
+## Intended uses & limitations
 
-# Load model weights
-state_dict = torch.load("pytorch_model.bin")
+More information needed
 
-# Initialize model (you'll need the BiEncoderModel class)
-model = BiEncoderModel(
-    base_model=AutoModel.from_pretrained("bert-base-uncased"),
-    num_classes=4
-)
-model.load_state_dict(state_dict)
-```
+## 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: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 7
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
+| 0.1571        | 1.0   | 78   | 0.1334          | 0.5761   | 0.5597 | 0.5846    | 0.5761 |
+| 0.1019        | 2.0   | 156  | 0.1009          | 0.6472   | 0.6399 | 0.6660    | 0.6472 |
+| 0.0711        | 3.0   | 234  | 0.0923          | 0.6861   | 0.6869 | 0.6910    | 0.6861 |
+| 0.0475        | 4.0   | 312  | 0.0972          | 0.6602   | 0.6611 | 0.6746    | 0.6602 |
+| 0.0258        | 5.0   | 390  | 0.1085          | 0.6602   | 0.6596 | 0.6699    | 0.6602 |
+| 0.0188        | 6.0   | 468  | 0.1099          | 0.6634   | 0.6626 | 0.6667    | 0.6634 |
+| 0.0155        | 7.0   | 546  | 0.1136          | 0.6602   | 0.6596 | 0.6642    | 0.6602 |
+
+
+### Framework versions
+
+- Transformers 4.45.1
+- Pytorch 2.4.0
+- Datasets 3.0.1
+- Tokenizers 0.20.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e0d720c8758edd4177b6e68903d9884cfabcf6a5642137d2cfe10329fbffcd7
+oid sha256:90f1f1375b8d1eda0e5d033ce68068eb4d8b3ccf89724de91e25df5e40f223f1
 size 437978280