Update README.md
Browse files
README.md
CHANGED
|
@@ -12,14 +12,17 @@ metrics:
|
|
| 12 |
model-index:
|
| 13 |
- name: QueryCategorizer
|
| 14 |
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
-
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
|
| 20 |
# QueryCategorizer
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
- Loss: 0.0038
|
| 25 |
- Accuracy: 0.9996
|
|
@@ -28,18 +31,32 @@ It achieves the following results on the evaluation set:
|
|
| 28 |
- F1: 0.9996
|
| 29 |
|
| 30 |
## Model description
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
|
| 32 |
-
More information needed
|
| 33 |
|
| 34 |
## Intended uses & limitations
|
| 35 |
|
| 36 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
|
| 38 |
## Training and evaluation data
|
| 39 |
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
## Training procedure
|
| 43 |
|
| 44 |
### Training hyperparameters
|
| 45 |
|
|
@@ -70,4 +87,4 @@ The following hyperparameters were used during training:
|
|
| 70 |
- Transformers 5.16.1
|
| 71 |
- Pytorch 2.11.0+cu128
|
| 72 |
- Datasets 4.8.5
|
| 73 |
-
- Tokenizers 0.23.1
|
|
|
|
| 12 |
model-index:
|
| 13 |
- name: QueryCategorizer
|
| 14 |
results: []
|
| 15 |
+
datasets:
|
| 16 |
+
- bitext/Bitext-customer-support-llm-chatbot-training-dataset
|
| 17 |
+
language:
|
| 18 |
+
- en
|
| 19 |
+
pipeline_tag: text-classification
|
| 20 |
---
|
| 21 |
|
|
|
|
|
|
|
| 22 |
|
| 23 |
# QueryCategorizer
|
| 24 |
|
| 25 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the bitext/Bitext-customer-support-llm-chatbot-training-dataset.
|
| 26 |
It achieves the following results on the evaluation set:
|
| 27 |
- Loss: 0.0038
|
| 28 |
- Accuracy: 0.9996
|
|
|
|
| 31 |
- F1: 0.9996
|
| 32 |
|
| 33 |
## Model description
|
| 34 |
+
A fine-tuned DistilBERT model that classifies customer support queries into one of 9 departments,
|
| 35 |
+
designed to auto-route incoming support tickets to the correct team.
|
| 36 |
+
## Model Details
|
| 37 |
+
|
| 38 |
+
- **Base model:** distilbert-base-uncased
|
| 39 |
+
- **Task:** Multi-class text classification (9 classes)
|
| 40 |
+
- **Language:** English
|
| 41 |
+
- **Fine-tuned on:** bitext/Bitext-customer-support-llm-chatbot-training-dataset
|
| 42 |
+
|
| 43 |
+
The model predicts one of the following 9 departments:
|
| 44 |
+
|
| 45 |
+
ACCOUNT, CONTACT, DELIVERY, FEEDBACK, INVOICE, ORDER, PAYMENT, REFUND, SHIPPING
|
| 46 |
|
|
|
|
| 47 |
|
| 48 |
## Intended uses & limitations
|
| 49 |
|
| 50 |
+
- Routing incoming customer support emails/messages to the appropriate department
|
| 51 |
+
- Prototyping multi-class text classification pipelines
|
| 52 |
+
- Educational demonstrations of fine-tuning transformer models for classification
|
| 53 |
+
- It is limited to just classify the message but not forward them to the appropriate department
|
| 54 |
+
- Out of 11 labels mentioned in dataset, we only took 9 and removed "Cancel" and "Subscription" because of limited no of data samples for that class.
|
| 55 |
|
| 56 |
## Training and evaluation data
|
| 57 |
|
| 58 |
+
The bitext/Bitext-customer-support-llm-chatbot-training-dataset was used for training and evaluation of the model.
|
| 59 |
+
80% was reserved for training, 10% for validation and 10% for testing.
|
|
|
|
| 60 |
|
| 61 |
### Training hyperparameters
|
| 62 |
|
|
|
|
| 87 |
- Transformers 5.16.1
|
| 88 |
- Pytorch 2.11.0+cu128
|
| 89 |
- Datasets 4.8.5
|
| 90 |
+
- Tokenizers 0.23.1
|