DistilBERT multi-label HR classifier v2

README.md

@@ -4,7 +4,6 @@ license: apache-2.0
 base_model: distilbert/distilbert-base-uncased
 tags:
 - generated_from_trainer
-- ml-intern
 metrics:
 - precision
 - recall
@@ -21,13 +20,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6631
-- F1 Micro: 0.1481
-- F1 Macro: 0.1452
-- Precision: 0.08
-- Recall: 1.0
+- Loss: 0.6809
+- F1 Micro: 0.1111
+- F1 Macro: 0.0470
+- Precision: 0.0714
+- Recall: 0.25
 - Accuracy: 0.0
-- Hamming: 0.92
+- Hamming: 0.32
 
 ## Model description
 
@@ -46,7 +45,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 3e-05
+- learning_rate: 1.0000000000000002e-06
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
@@ -55,16 +54,14 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 5
-- num_epochs: 8
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | Precision | Recall | Accuracy | Hamming |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------:|:------:|:--------:|:-------:|
-| 1.3535        | 1.0   | 5    | 0.6631          | 0.1481   | 0.1452   | 0.08      | 1.0    | 0.0      | 0.92    |
-| 1.1928        | 2.0   | 10   | 0.5828          | 0.1481   | 0.1452   | 0.08      | 1.0    | 0.0      | 0.92    |
-| 1.0551        | 3.0   | 15   | 0.5213          | 0.1401   | 0.1322   | 0.0759    | 0.9062 | 0.0      | 0.89    |
-| 0.9999        | 4.0   | 20   | 0.4835          | 0.1418   | 0.1274   | 0.0771    | 0.875  | 0.0      | 0.8475  |
+| 1.3633        | 1.0   | 5    | 0.6809          | 0.1111   | 0.0470   | 0.0714    | 0.25   | 0.0      | 0.32    |
+| 1.3502        | 2.0   | 10   | 0.6771          | 0.1      | 0.0450   | 0.0648    | 0.2188 | 0.0      | 0.315   |
 
 
 ### Framework versions
@@ -73,23 +70,3 @@ The following hyperparameters were used during training:
 - Pytorch 2.11.0+cu130
 - Datasets 4.8.5
 - Tokenizers 0.22.2
-
-<!-- ml-intern-provenance -->
-## Generated by ML Intern
-
-This model repository was generated by [ML Intern](https://github.com/huggingface/ml-intern), an agent for machine learning research and development on the Hugging Face Hub.
-
-- Try ML Intern: https://smolagents-ml-intern.hf.space
-- Source code: https://github.com/huggingface/ml-intern
-
-## Usage
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model_id = 'AurelPx/hr-conversations-classifier'
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id)
-```
-
-For non-causal architectures, replace `AutoModelForCausalLM` with the appropriate `AutoModel` class.

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47236dfdfc9cfc4c87c96dfdefc072d83982c74151298784f3a5007d77c165bb
+oid sha256:f756c07a9328f8dcd94706b0fe1b9b595c1db841a595c129037c4798cba79e0a
 size 267887936

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2a67b9b3a81433171067eee5ac5567f0effae5b3d842161da65bc3dda49ab2f
+oid sha256:c78eb30eb8bb36c69819fe090ac512e14369aeaef61d0a5f1f565ecc8296a795
 size 5329