Upload folder using huggingface_hub

README.md

@@ -1,3 +1,33 @@
 ---
+tags:
+  - text-to-image
+  - diffusion
+  - latent-diffusion
+  - pytorch
 license: apache-2.0
 ---
+
+# Mini Latent Diffusion Model
+
+A ~46M parameter UNet trained from scratch for text-to-image generation
+in latent space (SD VAE + CLIP text encoder, both frozen).
+
+**Dataset:** [BitTranslate/Bittensor_subnet_19_06_04_24](https://huggingface.co/datasets/BitTranslate/Bittensor_subnet_19_06_04_24)  
+**Trained steps:** 500  
+**Image size:** 256px → 32×32 latents
+
+## Architecture
+
+- **UNet base channels:** 128
+- **Channel multipliers:** (1, 2, 3, 4)
+- **Res blocks per level:** 2
+- **Cross-attention (text conditioning):** latent res (8,)
+- **VAE (frozen):** `runwayml/stable-diffusion-v1-5` vae subfolder
+- **Text encoder (frozen):** CLIP ViT-L/14 from `runwayml/stable-diffusion-v1-5`
+
+## Inference
+
+```python
+from train_mini_ldm import generate
+imgs = generate("a sunset over the ocean", "./mini_ldm_output/final")
+```

config.json

@@ -0,0 +1,27 @@
+{
+  "_model_type": "MiniUNet",
+  "in_channels": 4,
+  "model_channels": 128,
+  "channel_mults": [
+    1,
+    2,
+    3,
+    4
+  ],
+  "num_res_blocks": 2,
+  "attn_resolutions": [
+    8
+  ],
+  "context_dim": 768,
+  "dataset": "BitTranslate/Bittensor_subnet_19_06_04_24",
+  "image_size": 256,
+  "num_train_timesteps": 1000,
+  "beta_schedule": "scaled_linear",
+  "trained_steps": 500,
+  "vae_id": "runwayml/stable-diffusion-v1-5",
+  "vae_subfolder": "vae",
+  "text_encoder_id": "runwayml/stable-diffusion-v1-5",
+  "text_encoder_subfolder": "text_encoder",
+  "tokenizer_id": "runwayml/stable-diffusion-v1-5",
+  "tokenizer_subfolder": "tokenizer"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ea659a2e74cab259cf0b00737281940cf1ff3579437346d758f42a02be8c30f
+size 182146728

scheduler/scheduler_config.json

@@ -0,0 +1,19 @@
+{
+  "_class_name": "DDPMScheduler",
+  "_diffusers_version": "0.38.0",
+  "beta_end": 0.02,
+  "beta_schedule": "scaled_linear",
+  "beta_start": 0.0001,
+  "clip_sample": true,
+  "clip_sample_range": 1.0,
+  "dynamic_thresholding_ratio": 0.995,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "rescale_betas_zero_snr": false,
+  "sample_max_value": 1.0,
+  "steps_offset": 0,
+  "thresholding": false,
+  "timestep_spacing": "leading",
+  "trained_betas": null,
+  "variance_type": "fixed_small"
+}