Update generative-proof-of-concept-CPU-preprocessing-in-memory.py

david-thrower · web-flow · commit 61aca30a8f7e · 2025-10-14T23:29:30.000-04:00
Try to train the model directly on the CerebrosNotGPT wrapper.
diff --git a/generative-proof-of-concept-CPU-preprocessing-in-memory.py b/generative-proof-of-concept-CPU-preprocessing-in-memory.py
@@ -1407,7 +1407,8 @@ def create_dataset(raw_text_samples, tokenizer, sample_expansion_batch_size=10)
 
 
         phase_i_b_history =\
-                best_model_found.fit(
+                generator.model.fit(
+                   # best_model_found.fit(
                    x=phase_i_b_dataset,
                    epochs=phase_i_b_epochs,
                    batch_size=batch_size) # ,
