End of training

README.md

@@ -13,29 +13,37 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ProGemma2
 
-This is a custom configuration (336M parameters) of Google’s Gemma 2 LLM that is being pre-trained on amino acid sequences of 512 AA or less in length. Periodic updates are made to this page as training reaches new checkpoints.
+This model is a fine-tuned version of [JuIm/ProGemma2](https://huggingface.co/JuIm/ProGemma2) on an unknown dataset.
 
-The purpose of this model was to investigate the differences between ProGemma and ProtGPT (GPT-2 architecture) as it pertains to sequence generation. Training loss is ~1.5. Perplexity scores as well as AlphaFold 3’s ptm, pLDDT, and iptm scores are generally in line with ProtGPT’s scores for sequence lengths < 250, although the testing phase is still very early. I have yet to do testing for sequence lengths > 250. More robust testing is also required for lengths < 250 AA. In my very preliminary testing, HHblit e-values of ~0.1 are achieved with relatively easily.
+## Model description
 
-Controlled generation is not a capability of this model, and therefore serves as a method to significantly improve generation as, in principal, a sequence that performs a given function or resides in a particular cellular location can be generated.
+More information needed
 
-In sequence generation, a top_k of 950 appears to work well as it prevents repetition. This is also seen in ProtGPT.
+## Intended uses & limitations
 
-Below is code using the Transformers library to generate sequences using ProGemma.
+More information needed
 
-from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
+## Training and evaluation data
 
-model = AutoModelForCausalLM.from_pretrained("JuIm/ProGemma")
+More information needed
 
-tokenizer = AutoTokenizer.from_pretrained("JuIm/Amino-Acid-Sequence-Tokenizer")
+## Training procedure
 
-progemma = pipeline("text-generation", model=model, tokenizer=tokenizer)
+### Training hyperparameters
 
-sequence = progemma("\<bos>", top_k=950, max_length=100, num_return_sequences=1, do_sample=True, repetition_penalty=1.2, eos_token_id=21, pad_token_id=22, bos_token_id=20)
+The following hyperparameters were used during training:
+- learning_rate: 0.001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.4
+- training_steps: 3500
+
+### Training results
 
-s = sequence[0]['generated_text']
 
-print(s)
 
 ### Framework versions
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17b864d7750e05a521e6813a4a75f2bb4f74dd39bb39200a4c461306e96aa87a
+oid sha256:11cf34c26da8ccac37184ba87a4c736e0cd33cfe0dc50ed62896105db8148ec6
 size 1342562152

runs/Sep12_11-50-20_2ca4d4e85e16/events.out.tfevents.1726141826.2ca4d4e85e16.1458.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ba4e04df29d1e3c4e8dc62ffffd44840218bb4e739d561036c43c0575748cfa
+size 743400

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5413548608d28445ed2c278880efd16df5afffa5231bf3e127f1dfe5c5a69dc5
+oid sha256:4398007bb902d5ed7ea1d927eda96c8ea386c682a4e599c19bd2b3561e42b9a4
 size 5176