Upload trained VideoMAE with metrics and all figures

.gitattributes

@@ -35,3 +35,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 sample.gif filter=lfs diff=lfs merge=lfs -text
 longtail_f1_vs_freq.png filter=lfs diff=lfs merge=lfs -text
+confusion_matrix_test_50.png filter=lfs diff=lfs merge=lfs -text
+confusion_matrix_train_50.png filter=lfs diff=lfs merge=lfs -text
+confusion_matrix_valid_50.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -3,6 +3,8 @@ license: cc-by-nc-4.0
 base_model: MCG-NJU/videomae-base
 tags:
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: working
   results: []
@@ -14,6 +16,17 @@ should probably proofread and complete it, then remove this comment. -->
 # working
 
 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: 4.6095
+- Accuracy: 0.0266
+- Top 1 Accuracy: 0.0266
+- Top 5 Accuracy: 0.0858
+- Top 10 Accuracy: 0.1420
+- Macro Precision: 0.0024
+- Macro Recall: 0.0173
+- Macro F1: 0.0039
+- Pearson Corr: 0.3016
+- Spearman Corr: 0.2739
 
 ## Model description
 
@@ -41,11 +54,14 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 85
+- training_steps: 900
 - mixed_precision_training: Native AMP
 
 ### Training results
 
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Top 1 Accuracy | Top 5 Accuracy | Top 10 Accuracy | Macro Precision | Macro Recall | Macro F1 | Pearson Corr | Spearman Corr |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:--------------:|:--------------:|:---------------:|:---------------:|:------------:|:--------:|:------------:|:-------------:|
+| 4.523         | 5.0542 | 500  | 4.6095          | 0.0266   | 0.0266         | 0.0858         | 0.1420          | 0.0024          | 0.0173       | 0.0039   | 0.3016       | 0.2739        |
 
 
 ### Framework versions

config.json

@@ -21,7 +21,97 @@
     "6": "before",
     "7": "bird",
     "8": "birthday",
-    "9": "black"
+    "9": "black",
+    "10": "blue",
+    "11": "book",
+    "12": "bowling",
+    "13": "brown",
+    "14": "but",
+    "15": "can",
+    "16": "candy",
+    "17": "chair",
+    "18": "change",
+    "19": "cheat",
+    "20": "city",
+    "21": "clothes",
+    "22": "color",
+    "23": "computer",
+    "24": "cook",
+    "25": "cool",
+    "26": "corn",
+    "27": "cousin",
+    "28": "cow",
+    "29": "dance",
+    "30": "dark",
+    "31": "deaf",
+    "32": "decide",
+    "33": "doctor",
+    "34": "dog",
+    "35": "drink",
+    "36": "eat",
+    "37": "enjoy",
+    "38": "family",
+    "39": "fine",
+    "40": "finish",
+    "41": "fish",
+    "42": "forget",
+    "43": "full",
+    "44": "give",
+    "45": "go",
+    "46": "graduate",
+    "47": "hat",
+    "48": "hearing",
+    "49": "help",
+    "50": "hot",
+    "51": "how",
+    "52": "jacket",
+    "53": "kiss",
+    "54": "language",
+    "55": "last",
+    "56": "later",
+    "57": "letter",
+    "58": "like",
+    "59": "man",
+    "60": "many",
+    "61": "medicine",
+    "62": "meet",
+    "63": "mother",
+    "64": "need",
+    "65": "no",
+    "66": "now",
+    "67": "orange",
+    "68": "paint",
+    "69": "paper",
+    "70": "pink",
+    "71": "pizza",
+    "72": "play",
+    "73": "pull",
+    "74": "purple",
+    "75": "right",
+    "76": "same",
+    "77": "school",
+    "78": "secretary",
+    "79": "shirt",
+    "80": "short",
+    "81": "son",
+    "82": "study",
+    "83": "table",
+    "84": "tall",
+    "85": "tell",
+    "86": "thanksgiving",
+    "87": "thin",
+    "88": "thursday",
+    "89": "time",
+    "90": "walk",
+    "91": "want",
+    "92": "what",
+    "93": "white",
+    "94": "who",
+    "95": "woman",
+    "96": "work",
+    "97": "wrong",
+    "98": "year",
+    "99": "yes"
   },
   "image_size": 224,
   "initializer_range": 0.02,
@@ -36,7 +126,97 @@
     "before": 6,
     "bird": 7,
     "birthday": 8,
-    "black": 9
+    "black": 9,
+    "blue": 10,
+    "book": 11,
+    "bowling": 12,
+    "brown": 13,
+    "but": 14,
+    "can": 15,
+    "candy": 16,
+    "chair": 17,
+    "change": 18,
+    "cheat": 19,
+    "city": 20,
+    "clothes": 21,
+    "color": 22,
+    "computer": 23,
+    "cook": 24,
+    "cool": 25,
+    "corn": 26,
+    "cousin": 27,
+    "cow": 28,
+    "dance": 29,
+    "dark": 30,
+    "deaf": 31,
+    "decide": 32,
+    "doctor": 33,
+    "dog": 34,
+    "drink": 35,
+    "eat": 36,
+    "enjoy": 37,
+    "family": 38,
+    "fine": 39,
+    "finish": 40,
+    "fish": 41,
+    "forget": 42,
+    "full": 43,
+    "give": 44,
+    "go": 45,
+    "graduate": 46,
+    "hat": 47,
+    "hearing": 48,
+    "help": 49,
+    "hot": 50,
+    "how": 51,
+    "jacket": 52,
+    "kiss": 53,
+    "language": 54,
+    "last": 55,
+    "later": 56,
+    "letter": 57,
+    "like": 58,
+    "man": 59,
+    "many": 60,
+    "medicine": 61,
+    "meet": 62,
+    "mother": 63,
+    "need": 64,
+    "no": 65,
+    "now": 66,
+    "orange": 67,
+    "paint": 68,
+    "paper": 69,
+    "pink": 70,
+    "pizza": 71,
+    "play": 72,
+    "pull": 73,
+    "purple": 74,
+    "right": 75,
+    "same": 76,
+    "school": 77,
+    "secretary": 78,
+    "shirt": 79,
+    "short": 80,
+    "son": 81,
+    "study": 82,
+    "table": 83,
+    "tall": 84,
+    "tell": 85,
+    "thanksgiving": 86,
+    "thin": 87,
+    "thursday": 88,
+    "time": 89,
+    "walk": 90,
+    "want": 91,
+    "what": 92,
+    "white": 93,
+    "who": 94,
+    "woman": 95,
+    "work": 96,
+    "wrong": 97,
+    "year": 98,
+    "yes": 99
   },
   "layer_norm_eps": 1e-12,
   "model_type": "videomae",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f29e8222170b05063b362456cca791efbf5f84a159b1aa7224a2c4e51bed784
-size 344961984
+oid sha256:5cfb2a5d78e33cf5a29e338a7302154ff4740cbc75f20ae44937a5f1c93035a4
+size 345238832

per_class_report.csv

@@ -1,13 +1,103 @@
 ,precision,recall,f1-score,support,train_freq,pearson_corr,spearman_corr
-accident,0.0,0.0,0.0,3,13.0,0.24139568081361634,0.16353121952501118
-africa,0.0,0.0,0.0,3,12.0,0.24139568081361634,0.16353121952501118
-all,0.0,0.0,0.0,3,14.0,0.24139568081361634,0.16353121952501118
-apple,0.0,0.0,0.0,2,14.0,0.24139568081361634,0.16353121952501118
-basketball,0.2,0.5,0.28571428571428575,2,13.0,0.24139568081361634,0.16353121952501118
-bed,0.0,0.0,0.0,2,15.0,0.24139568081361634,0.16353121952501118
-before,0.1,0.5,0.16666666666666669,4,18.0,0.24139568081361634,0.16353121952501118
-bird,0.0,0.0,0.0,2,14.0,0.24139568081361634,0.16353121952501118
-birthday,0.0,0.0,0.0,3,12.0,0.24139568081361634,0.16353121952501118
-black,0.0,0.0,0.0,3,15.0,0.24139568081361634,0.16353121952501118
-macro avg,0.030000000000000006,0.1,0.045238095238095244,27,,0.24139568081361634,0.16353121952501118
-weighted avg,0.02962962962962963,0.1111111111111111,0.045855379188712526,27,,0.24139568081361634,0.16353121952501118
+accident,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+africa,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+all,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+apple,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+basketball,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+bed,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+before,0.0,0.0,0.0,4,4.0,0.2861190592244852,0.2664003109571473
+bird,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+birthday,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+black,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+blue,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+book,0.11764705882352941,1.0,0.21052631578947367,4,4.0,0.2861190592244852,0.2664003109571473
+bowling,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+brown,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+but,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+can,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+candy,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+chair,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+change,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+cheat,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+city,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+clothes,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+color,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+computer,0.0,0.0,0.0,5,5.0,0.2861190592244852,0.2664003109571473
+cook,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+cool,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+corn,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+cousin,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+cow,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+dance,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+dark,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+deaf,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+decide,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+doctor,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+dog,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+drink,0.025974025974025976,1.0,0.05063291139240507,4,4.0,0.2861190592244852,0.2664003109571473
+eat,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+enjoy,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+family,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+fine,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+finish,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+fish,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+forget,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+full,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+give,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+go,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+graduate,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+hat,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+hearing,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+help,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+hot,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+how,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+jacket,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+kiss,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+language,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+last,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+later,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+letter,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+like,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+man,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+many,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+medicine,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+meet,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+mother,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+need,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+no,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+now,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+orange,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+paint,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+paper,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+pink,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+pizza,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+play,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+pull,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+purple,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+right,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+same,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+school,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+secretary,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+shirt,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+short,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+son,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+study,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+table,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+tall,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+tell,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+thanksgiving,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+thin,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+thursday,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+time,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+walk,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+want,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+what,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+white,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+who,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+woman,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+work,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+wrong,0.0,0.0,0.0,2,2.0,0.2861190592244852,0.2664003109571473
+year,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+yes,0.0,0.0,0.0,3,3.0,0.2861190592244852,0.2664003109571473
+macro avg,0.001436210847975554,0.02,0.0026115922718187874,258,,0.2861190592244852,0.2664003109571473
+weighted avg,0.002226683485233417,0.031007751937984496,0.004048980266385717,258,,0.2861190592244852,0.2664003109571473

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9649a730047b49a30d9b1be48a42d1da0347c96566e7e7e958c5f579aa4f4b4f
+oid sha256:661255d78f9ac490d18693e6b65ed6a16745ded38bef5c9304ba0774aa2f5ffa
 size 4719