Skip to content

Updated metrics.py, exp_ForecastPFN.py, exp_transformer_metalearn.py, requirements.txt#9

Open
vaidatascientist wants to merge 1 commit intoabacusai:mainfrom
vaidatascientist:vai
Open

Updated metrics.py, exp_ForecastPFN.py, exp_transformer_metalearn.py, requirements.txt#9
vaidatascientist wants to merge 1 commit intoabacusai:mainfrom
vaidatascientist:vai

Conversation

@vaidatascientist
Copy link

A) requirements.txt:

  1. pmdarima==2.0.3 --> pmdarima==2.0.4
  2. tensorflow-io==0.26.0 --> tensorflow-io==0.37.1

B) ForecastPFN/benchmark/utils/metrics.py:

  1. from keras import backend --> from tensorflow.keras import backend

C) ForecastPFN/benchmark/exp/exp_ForecastPFN.py:

  1. Casted history variable to float32 to avoid error
  2. Used tf.saved_model.load() instead of tf.keras.models.load_model() as .pb models are not compatible with Keras3
  3. Added code to calculate SMAPE at the bottom

D) ForecastPFN/benchmark/exp/exp_transformer_metalearn.py

  1. Added comments near the imports where changing the sys.path.append() path is important to avoid local development errors

… requirements.txt

A) requirements.txt:
1. pmdarima==2.0.3 --> pmdarima==2.0.4
2. tensorflow-io==0.26.0 --> tensorflow-io==0.37.1

B) ForecastPFN/benchmark/utils/metrics.py:
1. from keras import backend --> from tensorflow.keras import backend

C) ForecastPFN/benchmark/exp/exp_ForecastPFN.py:
1. Casted history variable to float32 to avoid error
2. Used tf.saved_model.load() instead of tf.keras.models.load_model() as .pb models are not compatible with Keras3
3. Added code to calculate SMAPE at the bottom

D) ForecastPFN/benchmark/exp/exp_transformer_metalearn.py
1. Added comments near the imports where changing the sys.path.append() path is important to avoid local development errors
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant