Upload trained VideoMAE model with metrics and figures

README.md

@@ -5,9 +5,6 @@ tags:
 - generated_from_trainer
 metrics:
 - accuracy
-- precision
-- recall
-- f1
 model-index:
 - name: working
   results: []
@@ -20,14 +17,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3559
-- Top 1 Accuracy: 0.1562
-- Top 5 Accuracy: 0.5
+- Loss: 2.2936
+- Accuracy: 0.125
+- Top 1 Accuracy: 0.125
+- Top 5 Accuracy: 0.5938
 - Top 10 Accuracy: 1.0
-- Accuracy: 0.1562
-- Precision: 0.1367
-- Recall: 0.1562
-- F1: 0.0990
+- Macro Precision: 0.0129
+- Macro Recall: 0.1
+- Macro F1: 0.0229
 
 ## Model description
 
@@ -59,10 +56,10 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Top 1 Accuracy | Top 5 Accuracy | Top 10 Accuracy | Accuracy | Precision | Recall | F1     |
-|:-------------:|:------:|:----:|:---------------:|:--------------:|:--------------:|:---------------:|:--------:|:---------:|:------:|:------:|
-| No log        | 0.4706 | 8    | 2.3478          | 0.0312         | 0.4688         | 1.0             | 0.0312   | 0.0063    | 0.0312 | 0.0104 |
-| No log        | 1.4853 | 17   | 2.3559          | 0.1562         | 0.5            | 1.0             | 0.1562   | 0.1367    | 0.1562 | 0.0990 |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Top 1 Accuracy | Top 5 Accuracy | Top 10 Accuracy | Macro Precision | Macro Recall | Macro F1 |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:--------------:|:--------------:|:---------------:|:---------------:|:------------:|:--------:|
+| No log        | 0.4706 | 8    | 2.3049          | 0.125    | 0.125          | 0.625          | 1.0             | 0.0125          | 0.1          | 0.0222   |
+| No log        | 1.4853 | 17   | 2.2936          | 0.125    | 0.125          | 0.5938         | 1.0             | 0.0129          | 0.1          | 0.0229   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8768b95b8be983f716c0283a1781f80cc6c7ca066af2cb5672ed3bda23ac109e
+oid sha256:37e6425cc45ce3c51bb7cf2f09418d47c71b01584d38680b8e27a527861b1192
 size 344961984

per_class_report.csv

@@ -0,0 +1,14 @@
+,precision,recall,f1-score,support
+accident,0.0,0.0,0.0,3.0
+africa,0.0,0.0,0.0,3.0
+all,0.0,0.0,0.0,3.0
+apple,0.0,0.0,0.0,2.0
+basketball,0.0,0.0,0.0,2.0
+bed,0.0,0.0,0.0,2.0
+before,0.15384615384615385,1.0,0.2666666666666667,4.0
+bird,0.0,0.0,0.0,2.0
+birthday,0.0,0.0,0.0,3.0
+black,0.0,0.0,0.0,3.0
+accuracy,0.14814814814814814,0.14814814814814814,0.14814814814814814,0.14814814814814814
+macro avg,0.015384615384615385,0.1,0.026666666666666672,27.0
+weighted avg,0.022792022792022793,0.14814814814814814,0.03950617283950618,27.0

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcd09c72001d33e2207fee64ce97c506d3b244e526ec1acc22f19198fdddbbcf
+oid sha256:3386ccea732f35bbd068b84e8720d18f2b19f284ebbf7d37328e7a7404dafc2b
 size 4719