updated readme

.gitattributes

@@ -58,3 +58,18 @@ data/MIT_ast-finetuned-audioset-10-10-0-4593-embeddings_dev_train.npz filter=lfs
 data/MIT_ast-finetuned-audioset-10-10-0-4593-embeddings_dev_test.npz filter=lfs diff=lfs merge=lfs -text
 data/PCA_REDUCED_dcase2023_task2_baseline_ae-embeddings_dev_test.npz filter=lfs diff=lfs merge=lfs -text
 data/PCA_REDUCED_dcase2023_task2_baseline_ae-embeddings_dev_train.npz filter=lfs diff=lfs merge=lfs -text
+data/preview_dcase_2.png filter=lfs diff=lfs merge=lfs -text
+data/preview_dcase.png filter=lfs diff=lfs merge=lfs -text
+data/spotlight_save_layout.png filter=lfs diff=lfs merge=lfs -text
+data/config-spotlight-layout.json filter=lfs diff=lfs merge=lfs -text
+data/dcase2023_task2_baseline_ae-embeddings_dev_test.npz filter=lfs diff=lfs merge=lfs -text
+data/dev_test.tar.gz.lock filter=lfs diff=lfs merge=lfs -text
+data/dev_train.tar.gz filter=lfs diff=lfs merge=lfs -text
+data/preview_dcase_1.png filter=lfs diff=lfs merge=lfs -text
+data/dcase2023_task2_baseline_ae-embeddings_dev_train.npz filter=lfs diff=lfs merge=lfs -text
+data/dev_test.tar.gz filter=lfs diff=lfs merge=lfs -text
+data/dev_train.tar.gz.lock filter=lfs diff=lfs merge=lfs -text
+data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_add_train.npz filter=lfs diff=lfs merge=lfs -text
+data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_eval_test.npz filter=lfs diff=lfs merge=lfs -text
+data/add_train.tar.gz filter=lfs diff=lfs merge=lfs -text
+data/eval_test.tar.gz filter=lfs diff=lfs merge=lfs -text

7690148_urls.txt

@@ -0,0 +1,7 @@
+https://zenodo.org/record/7690148/files/dev_bearing.zip
+https://zenodo.org/record/7690148/files/dev_fan.zip
+https://zenodo.org/record/7690148/files/dev_gearbox.zip
+https://zenodo.org/record/7690148/files/dev_slider.zip
+https://zenodo.org/record/7690148/files/dev_ToyCar.zip
+https://zenodo.org/record/7690148/files/dev_ToyTrain.zip
+https://zenodo.org/record/7690148/files/dev_valve.zip

README.md

@@ -160,10 +160,6 @@ a ClassLabel for the label and a ClassLabel for the class.
  'embeddings_dcase2023_task2_baseline_ae': [12.602639198303223,
    16.997364044189453, ...,   
    -0.20931333303451538]
- 'anomaly_score_dcase2023_task2_baseline_ae': 8.284389
- 'prediction_dcase2023_task2_baseline_ae': 0
- 'prediction_correct_dcase2023_task2_baseline_ae': 1
- 'anomaly_score_embedding_lof': 0.191043
 }
 ```
 
@@ -183,11 +179,6 @@ The length of each audio file is 10 seconds.
 - `label`: an integer whose value may be either _0_, indicating that the audio sample is _normal_, _1_, indicating that the audio sample contains an _anomaly_.
 - `embeddings_ast-finetuned-audioset-10-10-0.4593`: an `datasets.Sequence(Value("float32"), shape=(1, 768))` representing audio embeddings that are generated with an [Audio Spectrogram Transformer](https://huggingface.co/docs/transformers/model_doc/audio-spectrogram-transformer#transformers.ASTFeatureExtractor).
 - `embeddings_dcase2023_task2_baseline_ae`: an `datasets.Sequence(Value("float32"), shape=(1, 512))` representing audio embeddings that are generated with the [**DCASE 2023 Challenge Task 2 Baseline Auto Encoder**](https://github.com/nttcslab/dcase2023_task2_baseline_ae). **Seven individual class-specific AEs** are trained. Dimensionality Reduction is applied with **PCA** separately for each class with a fit on the respecting training set of samples.
-- `anomaly_score_dcase2023_task2_baseline_ae`: a float representation of the anomaly score according to the baseline implementation
-- `prediction_dcase2023_task2_baseline_ae`: an integer whose value may be either _0_, indicating that the audio sample is considered _normal_ by the baseline algorithm, _1_, indicating that the audio sample contains an _anomaly_.
-- `prediction_correct_dcase2023_task2_baseline_ae`: an integer whose value may be either _0_, indicating that the baseline prediction is wrong or _1_, indicating that prediction is correct.
-- `anomaly_score_embedding_lof`: a float representation of the anomaly score computed with the PyOD implementation of the Local Outlier Factor algorithm on the pre-computed embedding.
-
 
 ### Data Splits
 
@@ -198,7 +189,18 @@ The development dataset has 2 splits: _train_ and _test_.
 | Train         | 7000                         | 6930 / 70                             |
 | Test          | 1400                         | 700 / 700                             |
 
-The information for the evaluation dataset will follow after release.
+The additional training dataset has 1 split: _train_.
+
+| Dataset Split | Number of Instances in Split | Source Domain / Target Domain Samples |
+| ------------- |------------------------------|---------------------------------------|
+| Train         | 7000                         | 6930 / 70                             |
+
+The evaluation dataset has 1 split: _test_.
+
+| Dataset Split | Number of Instances in Split | Source Domain / Target Domain Samples |
+|---------------|------------------------------|---------------------------------------|
+| Test          | 1400                         | ?                                     |
+
 
 ## Dataset Creation
 
@@ -300,22 +302,22 @@ If you use this dataset, please cite all the following papers. We will publish a
 - Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, and Masahiro Yasuda. First-shot anomaly detection for machine condition monitoring: a domain generalization baseline. In arXiv e-prints: 2303.00455, 2023. [[URL](https://arxiv.org/abs/2303.00455.pdf)]
 
 ```
-@dataset{kota_dohi_2023_7687464,
+@dataset{kota_dohi_2023_7882613,
   author       = {Kota Dohi and
-                  Keisuke and
-                  Noboru and
-                  Daisuke and
-                  Yuma and
-                  Tomoya and
-                  Harsh and
-                  Takashi and
-                  Yohei},
+                  Keisuke Imoto and
+                  Noboru Harada and
+                  Daisuke Niizumi and
+                  Yuma Koizumi and
+                  Tomoya Nishida and
+                  Harsh Purohit and
+                  Takashi Endo and
+                  Yohei Kawaguchi},
   title        = {DCASE 2023 Challenge Task 2 Development Dataset},
   month        = mar,
   year         = 2023,
   publisher    = {Zenodo},
-  version      = {1.0},
-  doi          = {10.5281/zenodo.7687464},
-  url          = {https://doi.org/10.5281/zenodo.7687464}
+  version      = {3.0},
+  doi          = {10.5281/zenodo.7882613},
+  url          = {https://doi.org/10.5281/zenodo.7882613}
 }
 ```

dcase23-task2-enriched.py

@@ -13,23 +13,23 @@ from typing import Iterable, Dict, Optional, Union, List
 
 
 _CITATION = """\
-@dataset{kota_dohi_2023_7687464,
+@dataset{kota_dohi_2023_7882613,
   author       = {Kota Dohi and
-                  Keisuke and
-                  Noboru and
-                  Daisuke and
-                  Yuma and
-                  Tomoya and
-                  Harsh and
-                  Takashi and
-                  Yohei},
+                  Keisuke Imoto and
+                  Noboru Harada and
+                  Daisuke Niizumi and
+                  Yuma Koizumi and
+                  Tomoya Nishida and
+                  Harsh Purohit and
+                  Takashi Endo and
+                  Yohei Kawaguchi},
   title        = {DCASE 2023 Challenge Task 2 Development Dataset},
   month        = mar,
   year         = 2023,
   publisher    = {Zenodo},
-  version      = {1.0},
-  doi          = {10.5281/zenodo.7687464},
-  url          = {https://doi.org/10.5281/zenodo.7687464}
+  version      = {3.0},
+  doi          = {10.5281/zenodo.7882613},
+  url          = {https://doi.org/10.5281/zenodo.7882613}
 }
 """
 _LICENSE = "Creative Commons Attribution 4.0 International Public License"
@@ -37,13 +37,13 @@ _LICENSE = "Creative Commons Attribution 4.0 International Public License"
 _METADATA_REG = r"attributes_\d+.csv"
 
 _NUM_TARGETS = 2
-_NUM_CLASSES = 7
+_NUM_CLASSES = 14
 
 _TARGET_NAMES = ["normal", "anomaly"]
-_CLASS_NAMES = ["gearbox", "fan", "bearing", "slider", "ToyCar", "ToyTrain", "valve"]
+_CLASS_NAMES = ["gearbox", "fan", "bearing", "slider", "ToyCar", "ToyTrain", "valve", "bandsaw", "grinder", "shaker", "ToyDrone", "ToyNscale", "ToyTank", "Vacuum"]
 
 _HOMEPAGE = {
-    "dev": "https://zenodo.org/record/7687464#.Y_96q9LMLmH",
+    "dev": "https://zenodo.org/record/7690157",
     "add": "",
     "eval": "",
 }
@@ -52,24 +52,23 @@ DATA_URLS = {
     "dev": {
         "train": "data/dev_train.tar.gz",
         "test": "data/dev_test.tar.gz",
-        "metadata": "data/dev_metadata_extended.csv",
+        "metadata": "data/dev_metadata.csv",
     },
     "add":  {
         "train": "data/add_train.tar.gz",
-        "test": "data/add_test.tar.gz",
-        "metadata": "data/add_metadata_extended.csv",
+        "metadata": "data/add_metadata.csv",
     },
     "eval": {
         "test": "data/eval_test.tar.gz",
-        "metadata": "data/eval_metadata_extended.csv",
+        "metadata": None,
     },
 }
 
 EMBEDDING_URLS = {
     "dev": {
         "embeddings_ast-finetuned-audioset-10-10-0.4593": {
-            "train": "data/MIT_ast-finetuned-audioset-10-10-0-4593-embeddings_dev_train.npz",
-            "test": "data/MIT_ast-finetuned-audioset-10-10-0-4593-embeddings_dev_test.npz",
+            "train": "data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_dev_train.npz",
+            "test": "data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_dev_test.npz",
             "size": (1, 768),
             "dtype": "float32",
         },
@@ -83,22 +82,16 @@ EMBEDDING_URLS = {
     },
     "add":  {
         "embeddings_ast-finetuned-audioset-10-10-0.4593": {
-            "train": "",
-            "test": "",
-        },
-        "embeddings_dcase2023_task2_baseline_ae": {
-            "train": "",
-            "test": "",
+            "train": "data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_add_train.npz",
+            "size": (1, 768),
+            "dtype": "float32",
         },
     },
     "eval": {
         "embeddings_ast-finetuned-audioset-10-10-0.4593": {
-            "train": "",
-            "test": "",
-        },
-        "embeddings_dcase2023_task2_baseline_ae": {
-            "train": "",
-            "test": "",
+            "test": "data/MIT_ast-finetuned-audioset-10-10-0.4593-embeddings_eval_test.npz",
+            "size": (1, 768),
+            "dtype": "float32",
         },
     },
 }
@@ -108,22 +101,22 @@ STATS = {
     "configs": {
         'dev': {
             'date': "Mar 1, 2023",
-            'version': "1.0.0",
-            'homepage': "https://zenodo.org/record/7687464#.ZABmANLMLmH",
+            'version': "3.0.0",
+            'homepage': "https://zenodo.org/record/7882613",
             "splits": ["train", "test"],
         },
-        # 'add': {
-        #     'date': None,
-        #     'version': "0.0.0",
-        #     'homepage': None,
-        #     "splits": ["train", "test"],
-        # },
-        # 'eval': {
-        #     'date': None,
-        #     'version': "0.0.0",
-        #     'homepage': None,
-        #     "splits": ["test"],
-        # },
+        'add': {
+            'date': "Apr 15, 2023",
+            'version': "1.0.0",
+            'homepage': "https://zenodo.org/record/7830345",
+            "splits": ["train"],
+        },
+        'eval': {
+            'date': "May 1, 2023",
+            'version': "1.0.0",
+            'homepage': "https://zenodo.org/record/7860847",
+            "splits": ["test"],
+        },
     }
 }
 
@@ -381,10 +374,6 @@ class DCASE2023Task2Dataset(datasets.GeneratorBasedBuilder):
                     "d2v": datasets.Value("string"),
                     "d3p": datasets.Value("string"),
                     "d3v": datasets.Value("string"),
-                    "anomaly_score_dcase2023_task2_baseline_ae": datasets.Value("float32"),
-                    "prediction_dcase2023_task2_baseline_ae": datasets.Value("int64"),
-                    "prediction_correct_dcase2023_task2_baseline_ae": datasets.Value("int64"),
-                    "anomaly_score_embedding_lof": datasets.Value("float32"),
                 }
         if self.config.embeddings_urls is not None:
             features.update({
@@ -436,7 +425,7 @@ class DCASE2023Task2Dataset(datasets.GeneratorBasedBuilder):
                     "local_extracted_archive": local_extracted_archive[split],
                     "audio_files": dl_manager.iter_archive(audio_path[split]),
                     "embeddings": embeddings[split],
-                    "metadata_file": dl_manager.download_and_extract(self.config.data_urls["metadata"]),
+                    "metadata_file": dl_manager.download_and_extract(self.config.data_urls["metadata"]) if self.config.data_urls["metadata"] is not None else None,
                     "is_streaming": dl_manager.is_streaming,
                 },
             ) for split in split_type if split in self.config.splits
@@ -452,20 +441,22 @@ class DCASE2023Task2Dataset(datasets.GeneratorBasedBuilder):
         is_streaming: Optional[bool],
     ):
         """Yields examples."""
-        metadata = pd.read_csv(metadata_file)
+        if metadata_file is not None:
+            metadata = pd.read_csv(metadata_file)
         data_fields = list(self._info().features.keys())
 
         id_ = 0
         for path, f in audio_files:
             lookup = Path(path).parent.name + "/" + Path(path).name
-            if lookup in metadata["path"].values:
+            if metadata_file is None or lookup in metadata["path"].values:
                 path = os.path.join(local_extracted_archive, path) if local_extracted_archive else path
                 if is_streaming:
                     audio = {"path": path, "bytes": f.read()}
                 else:
                     audio = {"path": path, "bytes": None}
                 result = {field: None for field in data_fields}
-                result.update(metadata[metadata["path"] == lookup].T.squeeze().to_dict())
+                if metadata_file is not None:
+                    result.update(metadata[metadata["path"] == lookup].T.squeeze().to_dict())
                 for emb_key in embeddings.keys():
                     result[emb_key] = np.asarray(embeddings[emb_key][lookup]).squeeze().tolist()
                 result["path"] = path