Simplify card intro and drop the Source checkpoint column

README.md

@@ -11,32 +11,32 @@ tags:
 
 # CELL-FM weights
 
-Checkpoints behind the [CELL-FM CondenSeq demo](https://huggingface.co/spaces/BoHuangLab/CELL-FM).
+Checkpoints of CELL-FM.
 
-| File | Model | Source checkpoint |
-|---|---|---|
-| `condenseq/cellfm_seq2img.bin` | CELL-FM CS sequence-to-image generator (includes the ESM-C 600M encoder) | `pretrain_condenseq/cellfm_seq2img/checkpoint-50000` |
-| `condenseq/vae.bin` | Image VAE, 160x160, 3 down blocks, 4 latent channels | `pretrain_condenseq/vae/checkpoint-50000` |
-| `condenseq/vit_cls.bin` | ViT condensed/diffuse classifier, 2-channel 160x160 input | `PT_CondenSeq_img_ViT_cls_R1/checkpoint-10000` |
-| `condenseq/reference_nucleus.npy` | The DAPI channel every CondenSeq generation is conditioned on: `(1, 160, 160)` float32 in [-1, 1]. CondenSeq protein index 12626, image 0 — the same one the offline seq2img runs used | built |
-| `hpa/cellfm_seq2img.bin` | CELL-FM virtual-staining generator for HPA, 256x256, 3-channel conditioning (includes the ESM-C 600M encoder) | `pretrain_hpa/cellfm_seq2img/checkpoint-50000` |
-| `hpa/vae.bin` | Image VAE at 256x256, the one `hpa/cellfm_seq2img.bin` was trained against | `pretrain_hpa/vae/checkpoint-50000` |
-| `hpa/vit.bin` | Image-embedding ViT for HPA: 12 layers, 512 hidden, 2048 MLP, 8 heads, patch 4, 256x256 input. Trained to identify the protein in an image over 13,908 classes; the embedding is the representation, not the prediction. Four input channels — protein, nucleus, ER, microtubules | `pretrain_hpa/vit/checkpoint-16000` |
-| `hpa/cellfm_img2seq.bin` | CELL-FM image-to-sequence model for HPA, 512x512, 3-channel conditioning (includes the ESM-C 600M encoder) | `pretrain_hpa/cellfm_img2seq/checkpoint-60000` |
-| `hpa/vae_512.bin` | Image VAE at 512x512, the one `hpa/cellfm_img2seq.bin` was trained against — a different model from `hpa/vae.bin`, not a rename | `pretrain_hpa/vae_512/checkpoint-50000` |
-| `opencell/cellfm_vs.bin` | CELL-FM virtual-staining generator for OpenCell, 256x256, single nucleus conditioning channel, fine-tuned (includes the ESM-C 600M encoder) | `finetune_opencell/cellfm_vs/checkpoint-100000` |
-| `opencell/vae.bin` | Image VAE at 256x256, the one `opencell/cellfm_vs.bin` was trained against — OpenCell-finetuned, so not interchangeable with `hpa/vae.bin` despite the matching shape | `finetune_opencell/vae/checkpoint-50000` |
-| `opencell/vit.bin` | The same ViT fine-tuned on OpenCell, 1,311 classes. Identical backbone to `hpa/vit.bin`, but its input stem takes **two** channels — protein and nucleus — where the HPA one takes four, so the two cannot be swapped: the count is fixed in `conv_proj` and a mismatch fails there | `finetune_opencell/vit/checkpoint-10000` |
-| `opencell/vs_anchor_nucleus.npy` | The nucleus every OpenCell generation is conditioned on: `(1, 256, 256)` float32 in [-1, 1]. Gene ATG7, crop `CID001813_FID00035838_proj_11` — bit-for-bit the conditioning channel of the published offline run | built |
-| `opencell/vs_genes.csv` | OpenCell's 1,311 genes: name, protein name, UniProt accession, Ensembl id, localization annotation and sequence. The metadata table minus its image paths | built |
-| `opencell/vs_reference_cells.npz` | Four proteins in two matched pairs (POLR1A/SNRPF nuclear, LSM14A/DDX6 both P-body), each with one real OpenCell crop as a `(nucleus, protein)` float16 pair — the image shown beside the generated one | built |
-| `hpa/anchor_cell.npy` | The fixed cell every NLS-screening image is conditioned on: `(3, 256, 256)` float32 in [-1, 1], channels nucleus, ER, microtubules. HPA gene H3C13, cell crop `1194_B2_2_4` | built |
-| `hpa/anchor_masks.npz` | Two 256x256 boolean masks over that cell, `nucleus` and `cell`; cytoplasm is `cell & ~nucleus` | built |
-| `hpa/pls_anchor_nls.npz` | The cell PLS generation conditions on for nuclear signals: `cell` `(3, 512, 512)` nucleus/ER/microtubules and `protein` `(1, 512, 512)`, float32 in [-1, 1]. HPA gene PPM1G (Nucleoplasm), crop `392_B9_1_11` | built |
-| `hpa/pls_anchor_nes.npz` | The same for export signals. HPA gene DIAPH1 (Cytosol, Plasma membrane), crop `1608_B3_1_1` | built |
-| `hpa/proteome_aa_counts.json` | Residue counts over the 12,894 HPA proteins (7,940,784 residues), the proteome baseline the frequency analysis compares against | built |
-| `hpa/pls_reference_nls.csv` | The 320 published NLS signals: 20 independent draws at each of 16 tail lengths, 10-25 aa | `output/hpa/pls_generation/nls` |
-| `hpa/pls_reference_nes.csv` | The same 320 for export signals | `output/hpa/pls_generation/nes` |
+| File | Model |
+|---|---|
+| `condenseq/cellfm_seq2img.bin` | CELL-FM CS sequence-to-image generator (includes the ESM-C 600M encoder) |
+| `condenseq/vae.bin` | Image VAE, 160x160, 3 down blocks, 4 latent channels |
+| `condenseq/vit_cls.bin` | ViT condensed/diffuse classifier, 2-channel 160x160 input |
+| `condenseq/reference_nucleus.npy` | The DAPI channel every CondenSeq generation is conditioned on: `(1, 160, 160)` float32 in [-1, 1]. CondenSeq protein index 12626, image 0 — the same one the offline seq2img runs used |
+| `hpa/cellfm_seq2img.bin` | CELL-FM virtual-staining generator for HPA, 256x256, 3-channel conditioning (includes the ESM-C 600M encoder) |
+| `hpa/vae.bin` | Image VAE at 256x256, the one `hpa/cellfm_seq2img.bin` was trained against |
+| `hpa/vit.bin` | Image-embedding ViT for HPA: 12 layers, 512 hidden, 2048 MLP, 8 heads, patch 4, 256x256 input. Trained to identify the protein in an image over 13,908 classes; the embedding is the representation, not the prediction. Four input channels — protein, nucleus, ER, microtubules |
+| `hpa/cellfm_img2seq.bin` | CELL-FM image-to-sequence model for HPA, 512x512, 3-channel conditioning (includes the ESM-C 600M encoder) |
+| `hpa/vae_512.bin` | Image VAE at 512x512, the one `hpa/cellfm_img2seq.bin` was trained against — a different model from `hpa/vae.bin`, not a rename |
+| `opencell/cellfm_vs.bin` | CELL-FM virtual-staining generator for OpenCell, 256x256, single nucleus conditioning channel, fine-tuned (includes the ESM-C 600M encoder) |
+| `opencell/vae.bin` | Image VAE at 256x256, the one `opencell/cellfm_vs.bin` was trained against — OpenCell-finetuned, so not interchangeable with `hpa/vae.bin` despite the matching shape |
+| `opencell/vit.bin` | The same ViT fine-tuned on OpenCell, 1,311 classes. Identical backbone to `hpa/vit.bin`, but its input stem takes **two** channels — protein and nucleus — where the HPA one takes four, so the two cannot be swapped: the count is fixed in `conv_proj` and a mismatch fails there |
+| `opencell/vs_anchor_nucleus.npy` | The nucleus every OpenCell generation is conditioned on: `(1, 256, 256)` float32 in [-1, 1]. Gene ATG7, crop `CID001813_FID00035838_proj_11` — bit-for-bit the conditioning channel of the published offline run |
+| `opencell/vs_genes.csv` | OpenCell's 1,311 genes: name, protein name, UniProt accession, Ensembl id, localization annotation and sequence. The metadata table minus its image paths |
+| `opencell/vs_reference_cells.npz` | Four proteins in two matched pairs (POLR1A/SNRPF nuclear, LSM14A/DDX6 both P-body), each with one real OpenCell crop as a `(nucleus, protein)` float16 pair — the image shown beside the generated one |
+| `hpa/anchor_cell.npy` | The fixed cell every NLS-screening image is conditioned on: `(3, 256, 256)` float32 in [-1, 1], channels nucleus, ER, microtubules. HPA gene H3C13, cell crop `1194_B2_2_4` |
+| `hpa/anchor_masks.npz` | Two 256x256 boolean masks over that cell, `nucleus` and `cell`; cytoplasm is `cell & ~nucleus` |
+| `hpa/pls_anchor_nls.npz` | The cell PLS generation conditions on for nuclear signals: `cell` `(3, 512, 512)` nucleus/ER/microtubules and `protein` `(1, 512, 512)`, float32 in [-1, 1]. HPA gene PPM1G (Nucleoplasm), crop `392_B9_1_11` |
+| `hpa/pls_anchor_nes.npz` | The same for export signals. HPA gene DIAPH1 (Cytosol, Plasma membrane), crop `1608_B3_1_1` |
+| `hpa/proteome_aa_counts.json` | Residue counts over the 12,894 HPA proteins (7,940,784 residues), the proteome baseline the frequency analysis compares against |
+| `hpa/pls_reference_nls.csv` | The 320 published NLS signals: 20 independent draws at each of 16 tail lengths, 10-25 aa |
+| `hpa/pls_reference_nes.csv` | The same 320 for export signals |
 
 Hyperparameters for the CondenSeq models are set in `pipeline.py` in the Space and mirror
 `scripts/cell_fm_cs/evaluate_seq2img.sh` and