CNN trained on the MJSynth dataset. It's a dataset for text recognition on document images, synthetically generated, covering 90K English words.
Translator Train Vs Val Loss Graph (for translator1):
CNN trained on the MJSynth dataset. It's a dataset for text recognition on document images, synthetically generated, covering 90K English words.
Translator Train Vs Val Loss Graph (for translator1):