Merge pull request #9 from ChEB-AI/feature/api_downloadble_models

Fix - Api Models

Author: sfluegel05

README.md

@@ -8,9 +8,9 @@ A web application for the ensemble is available at https://chebifier.hastingslab
 
 Not all models can be installed automatically at the moment:
 - `chebai-graph` and its dependencies. To install them, follow
-the instructions in the [chebai-graph repository](https://github.com/ChEB-AI/python-chebai-graph). 
+the instructions in the [chebai-graph repository](https://github.com/ChEB-AI/python-chebai-graph).
 - `chemlog-extra` can be installed with `pip install git+https://github.com/ChEB-AI/chemlog-extra.git`
-- The automatically installed version of `c3p` may not work under Windows. If you want to run chebifier on Windows, we 
+- The automatically installed version of `c3p` may not work under Windows. If you want to run chebifier on Windows, we
 recommend using this forked version: `pip install git+https://github.com/sfluegel05/c3p.git`
 
 
@@ -38,11 +38,26 @@ The package provides a command-line interface (CLI) for making predictions using
 The ensemble configuration is given by a configuration file (by default, this is `chebifier/ensemble.yml`). If you
 want to change which models are included in the ensemble or how they are weighted, you can create your own configuration file.
 
-Model weights for deep learning models are downloaded automatically from [Hugging Face](https://huggingface.co/chebai).
+Model weights for deep learning models are automatically downloaded from [Hugging Face](https://huggingface.co/chebai).
+To use specific model weights from Hugging face, add the `load_model` key in your configuration file. For example:
+
+```yaml
+my_electra:
+  type: electra
+  load_model: "electra_chebi50_v241"
+```
+
+### Available model weights:
+
+* `electra_chebi50_v241`
+* `resgated_chebi50_v241`
+* `c3p_with_weights`
+
+
 However, you can also supply your own model checkpoints (see `configs/example_config.yml` for an example).
 
 ```bash
-# Make predictions 
+# Make predictions
 python -m chebifier predict --smiles "CC(=O)OC1=CC=CC=C1C(=O)O" --smiles "C1=CC=C(C=C1)C(=O)O"
 
 # Make predictions using SMILES from a file
@@ -96,7 +111,7 @@ Currently, the following models are supported:
 | `c3p` | A collection _Chemical Classifier Programs_, generated by LLMs based on the natural language definitions of ChEBI classes. | 338 | [Mungall, Christopher J., et al., 2025: Chemical classification program synthesis using generative artificial intelligence, arXiv](https://arxiv.org/abs/2505.18470) | [c3p](https://github.com/chemkg/c3p) |
 
 In addition, Chebifier also includes a ChEBI lookup that automatically retrieves the ChEBI superclasses for a class
-matched by a SMILES string. This is not activated by default, but can be included by adding 
+matched by a SMILES string. This is not activated by default, but can be included by adding
 ```yaml
 chebi_lookup:
     type: chebi_lookup
@@ -109,15 +124,15 @@ to your configuration file.
 
 Given a sample (i.e., a SMILES string) and models $m_1, m_2, \ldots, m_n$, the ensemble works as follows:
 1. Get predictions from each model $m_i$ for the sample.
-2. For each class $c$, aggregate predictions $p_c^{m_i}$ from all models that made a prediction for that class. 
+2. For each class $c$, aggregate predictions $p_c^{m_i}$ from all models that made a prediction for that class.
 The aggregation happens separately for all positive predictions (i.e., $p_c^{m_i} \geq 0.5$) and all negative predictions
 ($p_c^{m_i} < 0.5$). If the aggregated value is larger for the positive predictions than for the negative predictions,
 the ensemble makes a positive prediction for class $c$:
 
 <img width="2297" height="114" alt="image" src="https://github.com/user-attachments/assets/2f0263ae-83ac-41ea-938a-c71b46082c22" />
 <!-- For some reason, this formula does not render in GitHub markdown. Therefore, I rendered it locally and added it as an image. The rendered formula is:
 $$
-\text{ensemble}(c) = \begin{cases} 
+\text{ensemble}(c) = \begin{cases}
 1 & \text{if } \sum_{i: p_c^{m_i} \geq 0.5} [\text{confidence}_c^{m_i} \cdot \text{model_weight}_{m_i} \cdot \text{trust}_c^{m_i}] > \sum_{i: p_c^{m_i} < 0.5} [\text{confidence}_c^{m_i} \cdot \text{model_weight}_{m_i} \cdot \text{trust}_c^{m_i}] \\
 0 & \text{otherwise}
 \end{cases}
@@ -135,25 +150,25 @@ Therefore, if in doubt, we are more confident in the negative prediction.
 
 Confidence can be disabled by the `use_confidence` parameter of the predict method (default: True).
 
-The model_weight can be set for each model in the configuration file (default: 1). This is used to favor a certain 
-model independently of a given class. 
-Trust is based on the model's performance on a validation set. After training, we evaluate the Machine Learning models 
+The model_weight can be set for each model in the configuration file (default: 1). This is used to favor a certain
+model independently of a given class.
+Trust is based on the model's performance on a validation set. After training, we evaluate the Machine Learning models
 on a validation set for each class. If the `ensemble_type` is set to `wmv-f1`, the trust is calculated as 1 + the F1 score.
 If the `ensemble_type` is set to `mv` (the default), the trust is set to 1 for all models.
 
 ### Inconsistency resolution
-After a decision has been made for each class independently, the consistency of the predictions with regard to the ChEBI hierarchy 
+After a decision has been made for each class independently, the consistency of the predictions with regard to the ChEBI hierarchy
 and disjointness axioms is checked. This is
 done in 3 steps:
-- (1) First, the hierarchy is corrected. For each pair of classes $A$ and $B$ where $A$ is a subclass of $B$ (following 
-the is-a relation in ChEBI), we set the ensemble prediction of $B$ to 1 if the prediction of $A$ is 1. Intuitively 
+- (1) First, the hierarchy is corrected. For each pair of classes $A$ and $B$ where $A$ is a subclass of $B$ (following
+the is-a relation in ChEBI), we set the ensemble prediction of $B$ to 1 if the prediction of $A$ is 1. Intuitively
 speaking, if we have determined that a molecule belongs to a specific class (e.g., aromatic primary alcohol), it also
 belongs to the direct and indirect superclasses (e.g., primary alcohol, aromatic alcohol, alcohol).
 - (2) Next, we check for disjointness. This is not specified directly in ChEBI, but in an additional ChEBI module ([chebi-disjoints.owl](https://ftp.ebi.ac.uk/pub/databases/chebi/ontology/)).
 We have extracted these disjointness axioms into a CSV file and added some more disjointness axioms ourselves (see
 `data>disjoint_chebi.csv` and `data>disjoint_additional.csv`). If two classes $A$ and $B$ are disjoint and we predict
 both, we select one with the higher class score and set the other to 0.
-- (3) Since the second step might have introduced new inconsistencies into the hierarchy, we repeat the first step, but 
+- (3) Since the second step might have introduced new inconsistencies into the hierarchy, we repeat the first step, but
 with a small change. For a pair of classes $A \subseteq B$ with predictions $1$ and $0$, instead of setting $B$ to $1$,
 we now set $A$ to $0$. This has the advantage that we cannot introduce new disjointness-inconsistencies and don't have
 to repeat step 2.

chebifier/ensemble/base_ensemble.py

@@ -3,15 +3,15 @@
 
 import torch
 import tqdm
-from chebifier.inconsistency_resolution import PredictionSmoother
-from chebifier.utils import load_chebi_graph, get_disjoint_files
 
 from chebifier.check_env import check_package_installed
+from chebifier.hugging_face import download_model_files
+from chebifier.inconsistency_resolution import PredictionSmoother
 from chebifier.prediction_models.base_predictor import BasePredictor
+from chebifier.utils import get_disjoint_files, load_chebi_graph
 
 
 class BaseEnsemble:
-
     def __init__(
         self,
         model_configs: dict,
@@ -29,8 +29,6 @@ def __init__(
         for model_name, model_config in model_configs.items():
             model_cls = MODEL_TYPES[model_config["type"]]
             if "hugging_face" in model_config:
-                from chebifier.hugging_face import download_model_files
-
                 hugging_face_kwargs = download_model_files(model_config["hugging_face"])
             else:
                 hugging_face_kwargs = {}

configs/example_config.yml

@@ -1,6 +1,6 @@
 
 chemlog_peptides:
-    type: chemlog
+    type: chemlog_peptides
     model_weight: 100 # if chemlog is available, it always gets chosen
 my_resgated:
   type: resgated