Cielciel/aift-model-review-multiple-label-classification

README.md

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1712
-- Accuracy Thresh: 0.9398
+- Loss: 0.1798
+- Accuracy Thresh: 0.9327
 
 ## Model description
 
@@ -48,16 +48,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy Thresh |
 |:-------------:|:-----:|:----:|:---------------:|:---------------:|
-| No log        | 1.0   | 70   | 0.1784          | 0.9347          |
-| No log        | 2.0   | 140  | 0.1673          | 0.9398          |
-| No log        | 3.0   | 210  | 0.1723          | 0.9367          |
-| No log        | 4.0   | 280  | 0.1647          | 0.9408          |
-| No log        | 5.0   | 350  | 0.1679          | 0.9378          |
-| No log        | 6.0   | 420  | 0.1686          | 0.9388          |
-| No log        | 7.0   | 490  | 0.1676          | 0.9439          |
-| 0.0662        | 8.0   | 560  | 0.1727          | 0.9398          |
-| 0.0662        | 9.0   | 630  | 0.1715          | 0.9398          |
-| 0.0662        | 10.0  | 700  | 0.1712          | 0.9398          |
+| No log        | 1.0   | 70   | 0.4152          | 0.8000          |
+| No log        | 2.0   | 140  | 0.3026          | 0.8929          |
+| No log        | 3.0   | 210  | 0.2406          | 0.9204          |
+| No log        | 4.0   | 280  | 0.2176          | 0.9214          |
+| No log        | 5.0   | 350  | 0.2038          | 0.9224          |
+| No log        | 6.0   | 420  | 0.1924          | 0.9337          |
+| No log        | 7.0   | 490  | 0.1900          | 0.9378          |
+| 0.2652        | 8.0   | 560  | 0.1868          | 0.9306          |
+| 0.2652        | 9.0   | 630  | 0.1804          | 0.9337          |
+| 0.2652        | 10.0  | 700  | 0.1798          | 0.9327          |
 
 
 ### Framework versions

config.json

@@ -32,6 +32,7 @@
   "n_heads": 12,
   "n_layers": 6,
   "pad_token_id": 0,
+  "problem_type": "multi_label_classification",
   "qa_dropout": 0.1,
   "seq_classif_dropout": 0.2,
   "sinusoidal_pos_embds": false,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4368d2b10cfc2b0cd90ade9932c8b920349c3ceb492409990b8420b3260fe1bf
+oid sha256:1e0601c465c2b89e2c72137736ee8a82835a3651c432dbf1e6017523f91d3b7f
 size 267847948

runs/Jan08_10-46-10_aift-review-classification-multiple-label/events.out.tfevents.1704710773.aift-review-classification-multiple-label

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8819bf0e242573216d4d287cf740e7fef4a1cfeae7871d63098b614cb103763d
-size 85
+oid sha256:6161846254307fd23700e04f86d555a225fbfc5d8e0a6ba3ccc6a93d04b1845c
+size 305

runs/Jan08_10-46-10_aift-review-classification-multiple-label/events.out.tfevents.1704710977.aift-review-classification-multiple-label

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fb53ec4fdc52c83e28ac68528a537a50fec6694ccdbf13858c19549dbfa5933
+size 8735

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99fefdff6ded43ad1f840c9883d61cf3c2318535e3bd04223d73b714e9759c0a
+oid sha256:c71d1b327e193b1ecfe34e0325e77aa329a377febe37e9ac69f02984a5ef6cd6
 size 4347