Upload folder using huggingface_hub

README.md

@@ -1,3 +1,22 @@
----
-license: mit
----
+---
+library_name: perturblab
+tags:
+- biology
+- genomics
+- scgpt
+license: mit
+base_model: bowang-lab/scGPT
+---
+
+# scgpt-continual-pretrained
+
+## Model Description
+This model weights were originally downloaded from the [bowang-lab/scGPT](https://github.com/bowang-lab/scGPT) repository.
+It has been re-uploaded here for ease of use with the `perturblab` library.
+
+## Source
+- **Original Repository**: [https://github.com/bowang-lab/scGPT](https://github.com/bowang-lab/scGPT)
+- **Paper**: [scGPT: Foundation Model for Single Cell Biology](https://www.nature.com/articles/s41592-024-02201-0)
+
+## Usage
+Intended for internal use with the PerturbLab framework.

args.json

@@ -0,0 +1,56 @@
+{
+  "data_source": "/scratch/ssd004/datasets/cellxgene/scb_strict/human",
+  "save_dir": "/scratch/ssd004/datasets/cellxgene/save/cellxgene_census_human-Dec18-13-52-2023",
+  "load_model": "/scratch/ssd004/datasets/cellxgene/save/scGPT_human",
+  "n_hvg": null,
+  "valid_size_or_ratio": 0.003,
+  "dist_backend": "nccl",
+  "grad_accu_steps": 1,
+  "pad_token": "<pad>",
+  "input_style": "binned",
+  "input_emb_style": "continuous",
+  "n_bins": 51,
+  "max_seq_len": 1200,
+  "training_tasks": "both",
+  "dist_url": "tcp://gpu183.cluster.local:54165",
+  "mask_ratio": [
+    0.25,
+    0.5,
+    0.75
+  ],
+  "trunc_by_sample": true,
+  "vocab_path": "/scratch/ssd004/datasets/cellxgene/scFormer/scformer/tokenizer/default_census_vocab.json",
+  "rank": 0,
+  "batch_size": 24,
+  "eval_batch_size": 48,
+  "epochs": 10,
+  "lr": 0.0001,
+  "scheduler_interval": 100,
+  "scheduler_factor": 0.99,
+  "warmup_ratio_or_step": 10000.0,
+  "no_cls": false,
+  "no_cce": true,
+  "fp16": true,
+  "fast_transformer": true,
+  "annotation_source": "/scratch/ssd004/datasets/cellxgene/tabula_sapiens/parquet/",
+  "annotation_valid_size_or_ratio": 0.1,
+  "nlayers": 12,
+  "nheads": 8,
+  "embsize": 512,
+  "d_hid": 512,
+  "dropout": 0.2,
+  "n_layers_cls": 3,
+  "annote_max_seq_len": 5000,
+  "log_interval": 500,
+  "save_interval": 1000,
+  "mask_value": -1,
+  "pad_value": -2,
+  "USE_CLS": true,
+  "USE_CCE": false,
+  "MVC": true,
+  "USE_GENERATIVE_TRAINING": true,
+  "world_size": 8,
+  "distributed": true,
+  "local_rank": 0,
+  "gpu": 0
+}

best_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad0252a1971e0cd619b7116dbab3177432236c4537225d54280a2aa7e5fe402a
+size 207861754