first version emodb dataset

Dockerfile

@@ -2,16 +2,16 @@ FROM python:3.9
 
 WORKDIR /code
 ENV HOME=/code
-COPY ./renumics_spotlight-1.3.0.post9+b32d1a8-py3-none-any.whl /code/renumics_spotlight-1.3.0.post9+b32d1a8-py3-none-any.whl
+COPY ./renumics_spotlight-1.3.0.post104+adfcaf1-py3-none-any /code/renumics_spotlight-1.3.0.post104+adfcaf1-py3-none-any
 # RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
 
 RUN apt install curl
 RUN pip install pip -U
 
-RUN pip install renumics_spotlight-1.3.0.post9+b32d1a8-py3-none-any.whl
+RUN pip install renumics_spotlight-1.3.0.post104+adfcaf1-py3-none-any
 
 
-RUN pip install pyarrow datasets cleanvision
+RUN pip install pyarrow datasets 
 
 COPY . .
 RUN mkdir -p /code/.cache

sliceguard-issues.pkl → emodb_embedding_issues.pickle

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe2c652c7600f011f0225e9e6fc9c133781409f7d1a1b4ce3a1c5871809cda25
-size 22835
+oid sha256:5f4a5b1d01f9e233e28b90b00cf5c797cc28b469ea8ee266c76db02ed436c44f
+size 1607

emodb_full_results.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18945df7fe2fd9a963ddc60295486bd77482a9202b66b3ae6ef7dec9e7d029a5
+size 11134181

cifar100-enrichment-cv.parquet → emodb_metadata_issues.pickle

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0bc7e67b598b765cce75309f7c414a13ca9fcc8004f436a0490822c899bfa66c
-size 544628
+oid sha256:6ae74f1f53f539a5ea1418210e8126bf6849b3b2a5366b170f9558893fca9953
+size 3695

layout-model-comparison.json

@@ -0,0 +1 @@
+{"orientation":"horizontal","children":[{"kind":"split","weight":22.443871668302805,"orientation":"vertical","children":[{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Table","type":"table","config":{"tableView":"selected","visibleColumns":null,"sorting":null,"orderByRelevance":false}}]},{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Similarity Map","type":"similaritymap","config":{"placeBy":null,"reductionMethod":null,"colorBy":"m1_gender_prediction_error","sizeBy":null,"filter":false,"umapNNeighbors":20,"umapMetric":null,"umapMinDist":0.15,"pcaNormalization":null,"umapMenuLocalGlobalBalance":null,"umapMenuIsAdvanced":false}}]}]},{"kind":"split","weight":19.08256196026494,"orientation":"vertical","children":[{"kind":"split","weight":4.406531194985268,"orientation":"horizontal","children":[{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Accuracy Filtered vs. Selected","type":"Metric","config":{"config":{"metric":"accuracy","columns":{"X":"m1_gender_prediction_error"}}}}]},{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"F1-Score Filtered vs. Selected","type":"Metric","config":{"config":{"metric":"F1","columns":{"X":"gender_bool","Y":"m1_gender_pred_bool"}}}}]}]},{"kind":"tab","weight":12.861339016257817,"children":[{"kind":"widget","name":"Confusion Matrix","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"m1_gender_prediction"}}]},{"kind":"tab","weight":12.861339016257817,"children":[{"kind":"widget","name":"Issues (from Sliceguard)","type":"IssuesWidget"}]}]},{"kind":"split","weight":17.45935648231709,"orientation":"vertical","children":[{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Gender vs. Emotion","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"emotion"}}]},{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Gender vs. Age","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"age"}}]}]},{"kind":"tab","weight":16.908583738243667,"children":[{"kind":"widget","name":"Inspector","type":"inspector","config":{"views":[{"view":"SpectrogramView","columns":["audio"],"name":"view","key":"07aaaf22-1586-453a-b9cb-674ea68f1079"},{"view":"AudioView","key":"5dKVmNM6c2fijG2GAPAKhD","name":"audio","columns":["audio"]},{"view":"ScalarView","columns":["gender"],"name":"view","key":"ac749b5f-98c5-47da-9026-df3134fd9ffb"},{"view":"TextLens","columns":["m1_gender_prediction_string"],"name":"view","key":"2c85598a-750e-4186-94aa-d311dd7aec37"}],"visibleColumns":4}}]}]}

layout-model-debug.json

@@ -0,0 +1 @@
+{"orientation":"horizontal","children":[{"kind":"split","weight":25.053244418990598,"orientation":"vertical","children":[{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Table","type":"table","config":{"tableView":"selected","visibleColumns":null,"sorting":null,"orderByRelevance":false}}]},{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Similarity Map","type":"similaritymap","config":{"placeBy":null,"reductionMethod":null,"colorBy":null,"sizeBy":null,"filter":false,"umapNNeighbors":20,"umapMetric":null,"umapMinDist":0.15,"pcaNormalization":null,"umapMenuLocalGlobalBalance":null,"umapMenuIsAdvanced":false}}]}]},{"kind":"split","weight":17.30115185099882,"orientation":"vertical","children":[{"kind":"tab","weight":4.406531194985268,"children":[{"kind":"widget","name":"Metric","type":"Metric","config":{"config":{"metric":"F1","columns":{"X":"m1_gender_prediction_error"}}}}]},{"kind":"tab","weight":25.722678032515635,"children":[{"kind":"widget","name":"Confusion Matrix","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"m1_gender_prediction"}}]}]},{"kind":"split","weight":16.905244485987403,"orientation":"vertical","children":[{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Confusion Matrix (3)","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"emotion"}}]},{"kind":"tab","weight":50,"children":[{"kind":"widget","name":"Confusion Matrix (2)","type":"ConfusionMatrix","config":{"xColumn":"gender","yColumn":"age"}}]}]},{"kind":"tab","weight":16.634733093151677,"children":[{"kind":"widget","name":"Inspector","type":"inspector","config":{"views":null,"visibleColumns":4}}]}]}

prepare.py

@@ -11,10 +11,10 @@ if __name__ == "__main__":
         print("Dataset loaded from cache.")
     else:
         # Load dataset using datasets.load_dataset()
-        dataset = datasets.load_dataset("renumics/cifar100-enriched", split="test")
+        dataset = datasets.load_dataset("renumics/emodb", split="test")
         print("Dataset loaded using datasets.load_dataset().")
         
-        df = dataset.to_pandas()  
+        df = dataset.to_pandas()      
 
         # Save dataset to cache
         df.to_parquet(cache_file)

renumics_spotlight-1.3.0.post9+b32d1a8-py3-none-any.whl → renumics_spotlight-1.3.0.post104+adfcaf1-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:39f23ad15279cfeebb210ccd9ca29dc30075d5f9b61934cb23b55d07207317a5
-size 2494250
+oid sha256:da1729500bbe890af38f8fd7161b1a693a4aac10800f6341fdbbcfed49df3178
+size 2525322

run.py

@@ -6,28 +6,28 @@ import pandas as pd
 
 if __name__ == "__main__":
     cache_file = "dataset_cache.parquet"
-    cache_file_enrichment="cifar100-enrichment-cv.parquet"
-    cache_file_issues="sliceguard-issues.pkl"
+    cache_file_enrichment="emodb_full_results.parquet"
+    cache_file_issues="emodb_metadata_issues.pkl"
     
     if os.path.exists(cache_file):
         # Load dataset from cache
-        df = pd.read_parquet(cache_file) 
+        df_db = pd.read_parquet(cache_file) 
         print("Dataset loaded from cache.")
     else:
         # Load dataset using datasets.load_dataset()
-        dataset = datasets.load_dataset("renumics/cifar100-enriched", split="test")
+        dataset = datasets.load_dataset("renumics/emodb", split="test")
         print("Dataset loaded using datasets.load_dataset().")
         
-        df = dataset.to_pandas()  
+        df_db = dataset.to_pandas()  
 
         # Save dataset to cache
-        df.to_parquet(cache_file)
+        df_db.to_parquet(cache_file)
 
         print("Dataset saved to cache.")
     
-    df_cv=pd.read_parquet(cache_file_enrichment) 
+    df=pd.read_parquet(cache_file_enrichment) 
     
-    df = pd.concat([df, df_cv], axis=1)
+    df['audio'] = df_db['audio']
      
     with open(cache_file_issues, "rb") as issue_file:
         issues = pickle.load(issue_file)
@@ -35,6 +35,10 @@ if __name__ == "__main__":
     #df = dataset.to_pandas()
     df_show = df.drop(columns=['embedding', 'probabilities'])
     while True:
-        view = spotlight.show(df_show, issues=issues, port=7860, host="0.0.0.0", layout="sliceguard-layout.json",
-                    dtype={"image": spotlight.Image, "embedding_reduced": spotlight.Embedding}, allow_filebrowsing=False)
+        dtypes = dtype={"audio": spotlight.Audio, 
+                'm1_embedding':spotlight.Embedding, 
+                'm2_embedding':spotlight.Embedding, 
+                'embedding_emotion':spotlight.Embedding}
+        view = spotlight.show(df_show, issues=issues, port=7860, host="0.0.0.0", layout="layout-model-debug.json",
+                    dtype=dtypes}, allow_filebrowsing=False)
         view.close()

sliceguard-layout.json

@@ -1 +0,0 @@
-{"orientation":"vertical","children":[{"kind":"split","weight":55.98243688254665,"orientation":"horizontal","children":[{"kind":"tab","weight":30.029340660631746,"children":[{"kind":"widget","name":"Table","type":"table","config":{"tableView":"full","visibleColumns":null,"sorting":null,"orderByRelevance":false}}]},{"kind":"tab","weight":27.735432693500154,"children":[{"kind":"widget","name":"Similarity Map","type":"similaritymap","config":{"placeBy":null,"reductionMethod":null,"colorBy":"fine_label_prediction_error","sizeBy":null,"filter":false,"umapNNeighbors":20,"umapMetric":null,"umapMinDist":0.15,"pcaNormalization":null,"umapMenuLocalGlobalBalance":null,"umapMenuIsAdvanced":false}}]},{"kind":"tab","weight":42.235226645868096,"children":[{"kind":"widget","name":"Issues","type":"IssuesWidget","config":null},{"kind":"widget","name":"Scatter Plot","type":"scatterplot","config":{"xAxisColumn":null,"yAxisColumn":null,"colorBy":null,"sizeBy":null,"filter":false}},{"kind":"widget","name":"Histogram","type":"histogram","config":{"columnKey":"dark_score","stackByColumnKey":"fine_label_prediction_error","filter":false}}]}]},{"kind":"tab","weight":44.01756311745335,"children":[{"kind":"widget","name":"Inspector","type":"inspector","config":{"views":[{"view":"ImageView","columns":["full_image"],"name":"full_image","key":"8GcjVfK27yKRQzznAqWMRG"},{"view":"TextLens","columns":["fine_label_prediction_str"],"name":"view","key":"797cd1ce-e56a-42de-87ac-f82d09a4f519"},{"view":"TextLens","columns":["fine_label_str"],"name":"view","key":"37df2c98-2676-4479-9984-8b30516fcf40"}],"visibleColumns":4}}]}]}