scene_segmentation

<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->

MariaK/scene_segmentation

This model is a fine-tuned version of nvidia/mit-b0 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 1.7868
• Validation Loss: 3.0539
• Validation Mean Iou: 0.0483
• Validation Mean Accuracy: 0.0908
• Validation Overall Accuracy: 0.3160
• Validation Accuracy Wall: 0.1166
• Validation Accuracy Building: 0.4198
• Validation Accuracy Sky: 0.1940
• Validation Accuracy Floor: 0.9531
• Validation Accuracy Tree: 0.3920
• Validation Accuracy Ceiling: 0.4672
• Validation Accuracy Road: 0.1731
• Validation Accuracy Bed : 0.3648
• Validation Accuracy Windowpane: 0.1441
• Validation Accuracy Grass: 0.0217
• Validation Accuracy Cabinet: 0.0857
• Validation Accuracy Sidewalk: nan
• Validation Accuracy Person: nan
• Validation Accuracy Earth: 0.0
• Validation Accuracy Door: 0.0
• Validation Accuracy Table: 0.0
• Validation Accuracy Mountain: 0.0
• Validation Accuracy Plant: 0.2078
• Validation Accuracy Curtain: 0.0
• Validation Accuracy Chair: 0.0
• Validation Accuracy Car: nan
• Validation Accuracy Water: 0.0
• Validation Accuracy Painting: 0.0
• Validation Accuracy Sofa: nan
• Validation Accuracy Shelf: nan
• Validation Accuracy House: nan
• Validation Accuracy Sea: nan
• Validation Accuracy Mirror: nan
• Validation Accuracy Rug: 0.0
• Validation Accuracy Field: 0.0
• Validation Accuracy Armchair: nan
• Validation Accuracy Seat: nan
• Validation Accuracy Fence: 0.0
• Validation Accuracy Desk: 0.0
• Validation Accuracy Rock: 0.0
• Validation Accuracy Wardrobe: 0.0
• Validation Accuracy Lamp: nan
• Validation Accuracy Bathtub: nan
• Validation Accuracy Railing: 0.0
• Validation Accuracy Cushion: nan
• Validation Accuracy Base: nan
• Validation Accuracy Box: nan
• Validation Accuracy Column: nan
• Validation Accuracy Signboard: nan
• Validation Accuracy Chest of drawers: 0.0
• Validation Accuracy Counter: 0.0
• Validation Accuracy Sand: nan
• Validation Accuracy Sink: nan
• Validation Accuracy Skyscraper: 0.0
• Validation Accuracy Fireplace: nan
• Validation Accuracy Refrigerator: nan
• Validation Accuracy Grandstand: nan
• Validation Accuracy Path: nan
• Validation Accuracy Stairs: 0.0
• Validation Accuracy Runway: 0.0
• Validation Accuracy Case: nan
• Validation Accuracy Pool table: nan
• Validation Accuracy Pillow: nan
• Validation Accuracy Screen door: nan
• Validation Accuracy Stairway: nan
• Validation Accuracy River: nan
• Validation Accuracy Bridge: nan
• Validation Accuracy Bookcase: nan
• Validation Accuracy Blind: nan
• Validation Accuracy Coffee table: nan
• Validation Accuracy Toilet: nan
• Validation Accuracy Flower: nan
• Validation Accuracy Book: 0.0
• Validation Accuracy Hill: nan
• Validation Accuracy Bench: 0.0
• Validation Accuracy Countertop: nan
• Validation Accuracy Stove: nan
• Validation Accuracy Palm: nan
• Validation Accuracy Kitchen island: nan
• Validation Accuracy Computer: nan
• Validation Accuracy Swivel chair: 0.0
• Validation Accuracy Boat: 0.0
• Validation Accuracy Bar: nan
• Validation Accuracy Arcade machine: nan
• Validation Accuracy Hovel: nan
• Validation Accuracy Bus: nan
• Validation Accuracy Towel: nan
• Validation Accuracy Light: 0.0
• Validation Accuracy Truck: 0.0
• Validation Accuracy Tower: 0.0
• Validation Accuracy Chandelier: nan
• Validation Accuracy Awning: nan
• Validation Accuracy Streetlight: nan
• Validation Accuracy Booth: nan
• Validation Accuracy Television receiver: nan
• Validation Accuracy Airplane: nan
• Validation Accuracy Dirt track: nan
• Validation Accuracy Apparel: nan
• Validation Accuracy Pole: nan
• Validation Accuracy Land: nan
• Validation Accuracy Bannister: nan
• Validation Accuracy Escalator: nan
• Validation Accuracy Ottoman: nan
• Validation Accuracy Bottle: nan
• Validation Accuracy Buffet: nan
• Validation Accuracy Poster: nan
• Validation Accuracy Stage: nan
• Validation Accuracy Van: nan
• Validation Accuracy Ship: nan
• Validation Accuracy Fountain: nan
• Validation Accuracy Conveyer belt: nan
• Validation Accuracy Canopy: nan
• Validation Accuracy Washer: nan
• Validation Accuracy Plaything: nan
• Validation Accuracy Swimming pool: nan
• Validation Accuracy Stool: nan
• Validation Accuracy Barrel: nan
• Validation Accuracy Basket: nan
• Validation Accuracy Waterfall: nan
• Validation Accuracy Tent: nan
• Validation Accuracy Bag: nan
• Validation Accuracy Minibike: nan
• Validation Accuracy Cradle: nan
• Validation Accuracy Oven: nan
• Validation Accuracy Ball: nan
• Validation Accuracy Food: nan
• Validation Accuracy Step: nan
• Validation Accuracy Tank: nan
• Validation Accuracy Trade name: nan
• Validation Accuracy Microwave: nan
• Validation Accuracy Pot: nan
• Validation Accuracy Animal: nan
• Validation Accuracy Bicycle: nan
• Validation Accuracy Lake: nan
• Validation Accuracy Dishwasher: nan
• Validation Accuracy Screen: nan
• Validation Accuracy Blanket: nan
• Validation Accuracy Sculpture: nan
• Validation Accuracy Hood: nan
• Validation Accuracy Sconce: nan
• Validation Accuracy Vase: nan
• Validation Accuracy Traffic light: nan
• Validation Accuracy Tray: nan
• Validation Accuracy Ashcan: nan
• Validation Accuracy Fan: nan
• Validation Accuracy Pier: nan
• Validation Accuracy Crt screen: nan
• Validation Accuracy Plate: nan
• Validation Accuracy Monitor: nan
• Validation Accuracy Bulletin board: nan
• Validation Accuracy Shower: nan
• Validation Accuracy Radiator: nan
• Validation Accuracy Glass: nan
• Validation Accuracy Clock: nan
• Validation Accuracy Flag: nan
• Validation Iou Wall: 0.0423
• Validation Iou Building: 0.2014
• Validation Iou Sky: 0.0801
• Validation Iou Floor: 0.6618
• Validation Iou Tree: 0.0735
• Validation Iou Ceiling: 0.2882
• Validation Iou Road: 0.1721
• Validation Iou Bed : 0.1926
• Validation Iou Windowpane: 0.0337
• Validation Iou Grass: 0.0128
• Validation Iou Cabinet: 0.0439
• Validation Iou Sidewalk: 0.0
• Validation Iou Person: 0.0
• Validation Iou Earth: 0.0
• Validation Iou Door: 0.0
• Validation Iou Table: 0.0
• Validation Iou Mountain: 0.0
• Validation Iou Plant: 0.1782
• Validation Iou Curtain: 0.0
• Validation Iou Chair: 0.0
• Validation Iou Car: nan
• Validation Iou Water: 0.0
• Validation Iou Painting: 0.0
• Validation Iou Sofa: nan
• Validation Iou Shelf: nan
• Validation Iou House: nan
• Validation Iou Sea: nan
• Validation Iou Mirror: nan
• Validation Iou Rug: 0.0
• Validation Iou Field: 0.0
• Validation Iou Armchair: nan
• Validation Iou Seat: nan
• Validation Iou Fence: 0.0
• Validation Iou Desk: 0.0
• Validation Iou Rock: 0.0
• Validation Iou Wardrobe: 0.0
• Validation Iou Lamp: nan
• Validation Iou Bathtub: nan
• Validation Iou Railing: 0.0
• Validation Iou Cushion: nan
• Validation Iou Base: nan
• Validation Iou Box: nan
• Validation Iou Column: nan
• Validation Iou Signboard: nan
• Validation Iou Chest of drawers: 0.0
• Validation Iou Counter: 0.0
• Validation Iou Sand: nan
• Validation Iou Sink: nan
• Validation Iou Skyscraper: 0.0
• Validation Iou Fireplace: nan
• Validation Iou Refrigerator: nan
• Validation Iou Grandstand: nan
• Validation Iou Path: nan
• Validation Iou Stairs: 0.0
• Validation Iou Runway: 0.0
• Validation Iou Case: nan
• Validation Iou Pool table: nan
• Validation Iou Pillow: nan
• Validation Iou Screen door: nan
• Validation Iou Stairway: nan
• Validation Iou River: nan
• Validation Iou Bridge: nan
• Validation Iou Bookcase: nan
• Validation Iou Blind: nan
• Validation Iou Coffee table: nan
• Validation Iou Toilet: nan
• Validation Iou Flower: nan
• Validation Iou Book: 0.0
• Validation Iou Hill: nan
• Validation Iou Bench: 0.0
• Validation Iou Countertop: nan
• Validation Iou Stove: nan
• Validation Iou Palm: nan
• Validation Iou Kitchen island: nan
• Validation Iou Computer: nan
• Validation Iou Swivel chair: 0.0
• Validation Iou Boat: 0.0
• Validation Iou Bar: nan
• Validation Iou Arcade machine: nan
• Validation Iou Hovel: nan
• Validation Iou Bus: nan
• Validation Iou Towel: nan
• Validation Iou Light: 0.0
• Validation Iou Truck: 0.0
• Validation Iou Tower: 0.0
• Validation Iou Chandelier: nan
• Validation Iou Awning: nan
• Validation Iou Streetlight: nan
• Validation Iou Booth: nan
• Validation Iou Television receiver: nan
• Validation Iou Airplane: nan
• Validation Iou Dirt track: nan
• Validation Iou Apparel: nan
• Validation Iou Pole: nan
• Validation Iou Land: nan
• Validation Iou Bannister: nan
• Validation Iou Escalator: nan
• Validation Iou Ottoman: nan
• Validation Iou Bottle: nan
• Validation Iou Buffet: nan
• Validation Iou Poster: nan
• Validation Iou Stage: nan
• Validation Iou Van: nan
• Validation Iou Ship: nan
• Validation Iou Fountain: nan
• Validation Iou Conveyer belt: nan
• Validation Iou Canopy: nan
• Validation Iou Washer: nan
• Validation Iou Plaything: nan
• Validation Iou Swimming pool: nan
• Validation Iou Stool: nan
• Validation Iou Barrel: nan
• Validation Iou Basket: nan
• Validation Iou Waterfall: nan
• Validation Iou Tent: nan
• Validation Iou Bag: nan
• Validation Iou Minibike: nan
• Validation Iou Cradle: nan
• Validation Iou Oven: nan
• Validation Iou Ball: nan
• Validation Iou Food: nan
• Validation Iou Step: nan
• Validation Iou Tank: nan
• Validation Iou Trade name: nan
• Validation Iou Microwave: nan
• Validation Iou Pot: nan
• Validation Iou Animal: nan
• Validation Iou Bicycle: nan
• Validation Iou Lake: nan
• Validation Iou Dishwasher: nan
• Validation Iou Screen: nan
• Validation Iou Blanket: nan
• Validation Iou Sculpture: nan
• Validation Iou Hood: nan
• Validation Iou Sconce: nan
• Validation Iou Vase: nan
• Validation Iou Traffic light: nan
• Validation Iou Tray: nan
• Validation Iou Ashcan: nan
• Validation Iou Fan: nan
• Validation Iou Pier: nan
• Validation Iou Crt screen: nan
• Validation Iou Plate: nan
• Validation Iou Monitor: nan
• Validation Iou Bulletin board: nan
• Validation Iou Shower: nan
• Validation Iou Radiator: nan
• Validation Iou Glass: nan
• Validation Iou Clock: nan
• Validation Iou Flag: nan
• Epoch: 49

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 6e-05, 'decay_steps': 2000, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

The model card is truncated. Read the rest on Hugging Face.