Downloads · 30 days
5
13% of all-time downloads
Jonny00/falcon-rw-1b-biome-definition
falcon-rw-1b-biome-definition is a machine learning model from Jonny00. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.
Part of bachelor thesis focusing on generating 3d terrain from text.<br Biome Definition is a json formatted string detailing the biomes and their features. <br Together with SD Heightmap Generation v3.0, 3d terrain c…
Downloads · 30 days
5
13% of all-time downloads
All-time downloads
38
Public
Repo size
50.4 MB
Likes
0
Public
Click a slice to open those files.
.safetensors12.6 MB · 100%
From the Hugging Face model README
Part of bachelor thesis focusing on generating 3d terrain from text.<br> Biome Definition is a json formatted string detailing the biomes and their features. <br> Together with SD Heightmap Generation v3.0, 3d terrain can be procedurally generated.
<!-- Provide a quick summary of what the model is/does. -->Example Inference:<br>
<prompt> Alpine mountains and a forest adorned with red flowers.
Result:<br>
<prompt> Alpine mountains and a forest adorned with red flowers.<br>
<biomes> [{'Entities': {'Boulders': [], 'BouldersSpawnrate': 0.0, 'Grass': [{'BaseTint': '#ff0000', 'Id': 'flowers'}], 'GrassSpawnrate': 0.0, 'Trees': [], 'TreesSpawnrate': 0.0}, 'Name': 'Red Flowers', 'SpawnCondition': {'HeightRange': [0.0, 0.1], 'SlopeRange': [0.0, 0.1]}, 'Texturing': {'GroundTexture': 'grass', 'GroundTextureTint': '#7cfc00', 'SlopeTexture': 'grass', 'SlopeTextureTint': '#7cfc00'}}, {'Entities': {'Boulders': [], 'BouldersSpawnrate': 0.0, 'Grass': [], 'GrassSpawnrate': 0.0, 'Trees': [{'BaseTint': '#ffffff', 'Id': 'tree_medium'}], 'TreesSpawnrate': 0.8}, 'Name': 'Mountain Range', 'SpawnCondition': {'HeightRange': [0.0, 0.7], 'SlopeRange': [0.0, 0.5]}, 'Texturing': {'GroundTexture':'stone', 'GroundTextureTint': '#ffffff', 'SlopeTexture':'stone_smooth', 'SlopeTextureTint': '#ffffff'}}]
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
Requires: ~8GB VRAM Training Loss: 2% Validation Loss: 4%
Model is biased to certain ids like:
Model also has trouble overadjusting tints, leading to oversaturated colorations. Model sometimes continues output. Just end prompt at "}}]". Model sometimes creates incomplete biome definitions or places entities on wrong biome.
<!-- This section is meant to convey both technical and sociotechnical limitations. -->Finetuned on custom dataset of size 538 json formatted biome definitions consisting of:
[More Information Needed]