renjithks
/

distilbert-expense-ner

@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [Geotrend/distilbert-base-en-fr-de-no-da-cased](https://huggingface.co/Geotrend/distilbert-base-en-fr-de-no-da-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2115
-- Precision: 0.9383
-- Recall: 0.9345
-- F1: 0.9364
-- Accuracy: 0.9565
 ## Model description
@@ -54,21 +54,21 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 1.0   | 100  | 0.3620          | 0.8830    | 0.8673 | 0.8751 | 0.9167   |
-| No log        | 2.0   | 200  | 0.2250          | 0.9176    | 0.9094 | 0.9134 | 0.9405   |
-| No log        | 3.0   | 300  | 0.2174          | 0.9234    | 0.9167 | 0.9200 | 0.9433   |
-| No log        | 4.0   | 400  | 0.2214          | 0.9213    | 0.9183 | 0.9198 | 0.9428   |
-| 0.3742        | 5.0   | 500  | 0.2099          | 0.9241    | 0.9264 | 0.9253 | 0.9474   |
-| 0.3742        | 6.0   | 600  | 0.1900          | 0.9412    | 0.9328 | 0.9370 | 0.9597   |
-| 0.3742        | 7.0   | 700  | 0.2174          | 0.9318    | 0.9280 | 0.9299 | 0.9515   |
-| 0.3742        | 8.0   | 800  | 0.1941          | 0.9360    | 0.9353 | 0.9357 | 0.9584   |
-| 0.3742        | 9.0   | 900  | 0.2157          | 0.9368    | 0.9361 | 0.9365 | 0.9565   |
-| 0.0315        | 10.0  | 1000 | 0.2115          | 0.9383    | 0.9345 | 0.9364 | 0.9565   |
 ### Framework versions
-- Transformers 4.19.1
-- Pytorch 1.11.0+cu113
-- Datasets 2.2.1
 - Tokenizers 0.12.1

 This model is a fine-tuned version of [Geotrend/distilbert-base-en-fr-de-no-da-cased](https://huggingface.co/Geotrend/distilbert-base-en-fr-de-no-da-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.1641
+- Precision: 0.9489
+- Recall: 0.9430
+- F1: 0.9459
+- Accuracy: 0.9721
 ## Model description
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| No log        | 1.0   | 112  | 0.1750          | 0.8720    | 0.8891 | 0.8805 | 0.9446   |
+| No log        | 2.0   | 224  | 0.1339          | 0.8804    | 0.9078 | 0.8939 | 0.9515   |
+| No log        | 3.0   | 336  | 0.1157          | 0.9315    | 0.9295 | 0.9305 | 0.9666   |
+| No log        | 4.0   | 448  | 0.1291          | 0.9269    | 0.9326 | 0.9298 | 0.9666   |
+| 0.2164        | 5.0   | 560  | 0.1400          | 0.9247    | 0.9285 | 0.9266 | 0.9666   |
+| 0.2164        | 6.0   | 672  | 0.1463          | 0.9376    | 0.9347 | 0.9362 | 0.9689   |
+| 0.2164        | 7.0   | 784  | 0.1463          | 0.9327    | 0.9337 | 0.9332 | 0.9694   |
+| 0.2164        | 8.0   | 896  | 0.1711          | 0.9376    | 0.9337 | 0.9356 | 0.9661   |
+| 0.0274        | 9.0   | 1008 | 0.1621          | 0.9421    | 0.9440 | 0.9431 | 0.9735   |
+| 0.0274        | 10.0  | 1120 | 0.1641          | 0.9489    | 0.9430 | 0.9459 | 0.9721   |
 ### Framework versions
+- Transformers 4.18.0
+- Pytorch 1.11.0
+- Datasets 2.1.0
 - Tokenizers 0.12.1