joethequant commited on
Commit
7a46da1
1 Parent(s): dfaaf0f

Update README.md

Browse files

Added additional Information

Files changed (1) hide show
  1. README.md +49 -0
README.md CHANGED
@@ -63,6 +63,55 @@ Performance and analytics:
63
  ## How to Use
64
  Instructions on how to use the model, including example prompts and API documentation, are available in the [Code Repository](https://github.com/joethequant/docker_streamlit_antibody_protein_generation).
65
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
  ## Limitations and Future Work
67
  - Predictions require experimental validation for practical use.
68
  - Future improvements will focus on incorporating diverse training data and enhancing prediction accuracy for the efficacy of generated antibodies.
 
63
  ## How to Use
64
  Instructions on how to use the model, including example prompts and API documentation, are available in the [Code Repository](https://github.com/joethequant/docker_streamlit_antibody_protein_generation).
65
 
66
+ ### Example Code
67
+
68
+ ```python
69
+ from models.progen.modeling_progen import ProGenForCausalLM
70
+ import torch
71
+ from tokenizers import Tokenizer
72
+ import json
73
+
74
+ # Define the model identifier from Hugging Face's model hub
75
+ model_path = 'AntibodyGeneration/fine-tuned-progen2-small'
76
+
77
+ # Load the model and tokenizer
78
+ model = ProGenForCausalLM.from_pretrained(model_path)
79
+ tokenizer = Tokenizer.from_file('tokenizer.json')
80
+
81
+ # Define your sequence and other parameters
82
+ target_sequence = 'MQIPQAPWPVVWAVLQLGWRPGWFLDSPDRPWNPPTFSPALLVVTEGDNATFTCSFSNTSESFVLNWYRMSPSNQTDKLAAFPEDRSQPGQDCRFRVTQLPNGRDFHMSVVRARRNDSGTYLCGAISLAPKAQIKESLRAELRVTERRAEVPTAHPSPSPRPAGQFQTLVVGVVGGLLGSLVLLVWVLAVICSRAARGTIGARRTGQPLKEDPSAVPVFSVDYGELDFQWREKTPEPPVPCVPEQTEYATIVFPSGMGTSSPARRGSADGPRSAQPLRPEDGHCSWPL'
83
+ number_of_sequences = 2
84
+
85
+ # Tokenize the sequence
86
+ tokenized_sequence = tokenizer(target_sequence, return_tensors="pt")
87
+
88
+ # Move model and tensors to CUDA if available
89
+ device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
90
+ model = model.to(device)
91
+ tokenized_sequence = tokenized_sequence.to(device)
92
+
93
+ # Generate sequences
94
+ with torch.no_grad():
95
+ output = model.generate(**tokenized_sequence, max_length=1024, pad_token_id=tokenizer.pad_token_id, do_sample=True, top_p=0.9, temperature=0.8, num_return_sequences=number_of_sequences)
96
+
97
+ # Decoding the output to get generated sequences
98
+ generated_sequences = [tokenizer.decode(output_seq, skip_special_tokens=True) for output_seq in output]
99
+ ```
100
+
101
+ ## Links:
102
+
103
+ - [Huggingface Model Repository](https://huggingface.co/AntibodyGeneration)
104
+ - [Web Demo](https://orca-app-ygzbp.ondigitalocean.app/Demo_Antibody_Generator)
105
+ - [OpenSource RunPod Severless Rest API](https://github.com/joethequant/docker_protein_generator)
106
+ - [The Code for this App](https://github.com/joethequant/docker_streamlit_antibody_protein_generation)
107
+
108
+ ## Additional Resources and Links
109
+ - [Progen Foundation Models](https://github.com/salesforce/progen)
110
+ - [ANARCI Github](https://github.com/oxpig/ANARCI)
111
+ - [ANARCI Webserver](http://opig.stats.ox.ac.uk/webapps/anarci/)
112
+ - [TAP: Therapeutic Antibody Profiler](https://opig.stats.ox.ac.uk/webapps/sabdab-sabpred/sabpred/tap)
113
+ - [ESM Fold](https://esmatlas.com/resources?action=fold)
114
+
115
  ## Limitations and Future Work
116
  - Predictions require experimental validation for practical use.
117
  - Future improvements will focus on incorporating diverse training data and enhancing prediction accuracy for the efficacy of generated antibodies.